Feb. 4, 2025

Episode 404: AI Agents and Their Impact on ERP Software and Business

Apple Podcasts podcast player icon Spotify podcast player icon Youtube Music podcast player icon RSS Feed podcast player icon Amazon Music podcast player icon Audible podcast player icon YouTube Channel podcast player icon
Apple Podcasts podcast player icon Spotify podcast player icon Youtube Music podcast player icon RSS Feed podcast player icon Amazon Music podcast player icon Audible podcast player icon YouTube Channel podcast player icon
Season 4 Business Central Partner

🎙️ In this episode of Dynamics Corner, Kris and Brad converse with recent 🎉 Microsoft MVP Sai Turlapati 🎉. Listen in as Sai shares insights on the evolution of AI, particularly in the enterprise sector, emphasizing Microsoft's significant role in AI adoption. He discusses the importance of prompting in AI interactions and the emerging concept of agents that can automate tasks.
 
🎧 Listen to hear more of the conversation about:
➡️ How prompting is crucial for effective AI interaction
➡️ The practical applications of AI agents in scheduling transformative impact of AI on time management, enterprise applications, and business workflows
➡️ How AI copilots can enhance productivity and efficiency in various industries
➡️ The importance of adapting to new technologies and the potential challenges businesses face in integrating AI solutions.

Send us a text

#MSDyn365BC #BusinessCentral #BC #DynamicsCorner

Follow Kris and Brad for more content:
https://matalino.io/bio
https://bprendergast.bio.link/

Chapters

00:00 - Enterprise AI and Microsoft Tools

11:49 - Harnessing AI and Natural Language Processing

24:10 - Automating Scheduling With AI Agents

37:05 - Future of Enterprise AI and Technology

41:14 - AI Adoption in ERP and CRM

53:51 - Exploring AI Applications Across Industries

59:14 - Navigating Microsoft's AI Ecosystem

Transcript
WEBVTT

00:00:00.521 --> 00:00:03.730
Welcome everyone to another episode of Dynamics Corner.

00:00:03.730 --> 00:00:11.833
Copilot, jack of all trades, master of none, but oftentimes better than a master of one.

00:00:11.833 --> 00:00:13.586
I'm your co-host, Chris.

00:00:14.560 --> 00:00:15.262
And this is Brad.

00:00:15.262 --> 00:00:19.233
This episode was recorded on January 29th 2025.

00:00:19.233 --> 00:00:23.231
Chris, Chris, Chris, that was a good little jingle.

00:00:23.231 --> 00:00:26.678
Did you use Copilot to write that, Chris?

00:00:26.698 --> 00:00:26.838
Chris.

00:00:26.858 --> 00:00:27.899
Chris, that was a good little jingle.

00:00:27.899 --> 00:00:25.097
Did you use Copilot to write that which part?

00:00:28.640 --> 00:00:31.745
No, I did not, I did not use Copilot for that, but you did say a comment.

00:00:31.745 --> 00:00:35.039
You used that term, jack of all trades, master of none.

00:00:35.039 --> 00:00:41.665
And then I realized there's actually a full quote and this was very fitting.

00:00:41.665 --> 00:00:44.853
When you said that, I was like, ah, very fitting, I like that.

00:00:44.853 --> 00:00:45.779
I was like, ah, very fitting, I like that.

00:00:46.100 --> 00:00:59.813
And that was fitting, because today we had the opportunity to dive deeper into this world of AI, which everyone seems to be talking about, and there's a lot of information to unravel, and there will be a lot of information to unravel in the future as well, too.

00:00:59.813 --> 00:01:17.156
With us today, we had the opportunity to speak with Sai Charlapati about Copilot, ai and many other things hey good morning, good morning.

00:01:17.156 --> 00:01:18.117
Hey, good morning.

00:01:18.537 --> 00:01:19.198
How are you?

00:01:19.218 --> 00:01:19.418
doing.

00:01:19.418 --> 00:01:23.227
Hey, good morning Chris, Good morning Greg, how are you guys?

00:01:23.569 --> 00:01:25.786
Very good, very well, very well, thank you.

00:01:25.786 --> 00:01:28.748
Thank you for taking the time to speak with us.

00:01:28.748 --> 00:01:30.387
Been looking forward to speaking with you.

00:01:32.081 --> 00:01:33.768
Yeah, thanks for inviting me.

00:01:33.768 --> 00:01:36.569
I heard a lot of episodes.

00:01:36.569 --> 00:01:43.263
I'm really interested to talk to you guys and learn so much from your podcast, I think today.

00:01:43.343 --> 00:01:57.466
We're interested in speaking with you and learning a lot from you, or hearing a lot from you about some popular topics that I see a lot of information on and you also share a lot of information about, which is exciting.

00:01:58.281 --> 00:02:04.727
I'm getting old, so it's all difficult for me and it's very difficult for me to keep up with everything that's going on.

00:02:04.727 --> 00:02:22.091
It seems that everything's accelerating quickly and I just can't keep up, but that's why we get to talk with people such as yourself to hopefully share some insights, to help us get a handle and a better understanding on some of the technology that is available to us.

00:02:22.091 --> 00:02:26.925
Before we get into the topic, would you mind telling everybody a little bit about yourself?

00:02:28.319 --> 00:02:31.830
Yeah, sure, my name is Sai Thirulapati.

00:02:31.830 --> 00:02:35.461
I am in the IT industry for the past almost 20 years.

00:02:35.461 --> 00:02:37.865
I saw the Y2K.

00:02:37.865 --> 00:02:49.770
During the time I was very young, fresh out of college, trying fresh out of the college, trying to understand that mainframe transition and other things.

00:02:49.770 --> 00:02:52.099
Then I saw mobile revolution, then cloud revolution.

00:02:52.099 --> 00:03:07.013
So there are these waves of technology revolutions that we saw and I was able to ride those waves and recently, for the past few years, I was very interested in the AI space.

00:03:07.500 --> 00:03:09.067
So I looked at the different.

00:03:09.067 --> 00:03:15.066
Who are the players in the AI space, especially enterprise AI.

00:03:15.066 --> 00:03:16.503
The enterprise AI.

00:03:16.503 --> 00:03:29.669
Microsoft, claudie, who is the Anthropic, is the company that creates this quality, like open AI is having charge upt.

00:03:29.669 --> 00:03:32.840
These are the players, especially in the b2c space.

00:03:32.840 --> 00:03:33.862
That's how I see it.

00:03:33.862 --> 00:03:50.072
In the b2b space, microsoft, amazon, oracle and, you know, google are the players, but predominantly I see Google with Gemini and Microsoft with their own Azure framework.

00:03:50.072 --> 00:04:06.788
They started with Azure with the backend, trying to talk to any LLMs, but finally they decided to just create a wrapper around it and explore the you know the LLMs that are being developed by other players.

00:04:06.979 --> 00:04:20.069
So that's how I got interested in this space and I feel like the first wave of the impact of AI is going to be in the enterprise side, at least on the customer service and sales.

00:04:20.069 --> 00:04:25.826
That's how I see it, because that's where there is a quick value that enterprises can see.

00:04:25.826 --> 00:04:31.892
So in that space I evaluated who are the top players in the CRM and customer service.

00:04:31.892 --> 00:04:35.990
Salesforce is one of the top players and Microsoft is another one.

00:04:35.990 --> 00:04:39.512
Hubspot is there, sage CRM is there.

00:04:39.512 --> 00:04:41.459
Those are very good players.

00:04:42.081 --> 00:04:49.887
So in that I looked at who can really help the enterprises who are having the end to end story.

00:04:49.887 --> 00:04:57.608
When I looked at it, microsoft is having the teams right Microsoft teams and Salesforce is having them.

00:04:57.608 --> 00:05:07.380
Slack that's the company that they bought, so those two are going to be really competing in that space for the AI to get the enterprise adoption.

00:05:07.380 --> 00:05:17.567
And one thing that Salesforce is not having especially lacking is the cloud story, whereas Microsoft is having the good cloud story.

00:05:17.567 --> 00:05:19.692
I looked at the Google.

00:05:19.692 --> 00:05:32.048
Google is, gcp is having cloud stories, so as Amazon, but they don't have the enterprise software such as Dynamics 365, erps or CRM, customer service and all those things.

00:05:32.048 --> 00:05:44.269
Then I felt like, okay, I am in my 40s, I feel like I need to bet on one of the real vendors who are going to take me to next 10 to 20 years.

00:05:44.269 --> 00:05:45.867
I looked at Microsoft.

00:05:46.199 --> 00:05:47.084
I feel like okay.

00:05:47.103 --> 00:05:47.646
Microsoft is having.

00:05:47.646 --> 00:05:51.016
You took your bet on Microsoft versus Google.

00:05:52.300 --> 00:05:56.101
Yes, because Google is not having Chris.

00:05:56.101 --> 00:05:59.531
Google is not having any ERP or CRM.

00:05:59.531 --> 00:06:03.803
They tried to buy the HubSpot but they withdrew that bid recently.

00:06:03.803 --> 00:06:07.968
To buy the HubSpot, but they withdrew that bid recently.

00:06:07.968 --> 00:06:17.276
So for any cloud vendors, for that fact, for any enterprise companies, to build this CRM and ERP systems, it's long, you know, it takes a long time.

00:06:17.276 --> 00:06:22.502
And also, the important thing is the user base.

00:06:22.502 --> 00:06:46.670
Right, they can build the efficient software solutions, but I feel like attracting the users is a difficult thing, so you're going with Microsoft for the.

00:06:46.475 --> 00:06:50.630
B2B enterprise company to have a larger adoption within the B2B space because of the exposure to businesses with the existing applications that they can build upon utilizing.

00:06:50.435 --> 00:06:50.776
AI.

00:06:50.776 --> 00:06:50.699
Well, you covered a lot.

00:06:50.699 --> 00:06:50.595
See, it's already a lot on there.

00:06:50.595 --> 00:06:53.343
We're just getting into it, man, I know.

00:06:53.363 --> 00:06:55.069
We're scratching the surface.

00:06:55.069 --> 00:06:55.610
We got into it.

00:06:55.610 --> 00:06:57.783
I'm still back at see, my mind is still processing.

00:06:57.783 --> 00:07:05.920
I'm still back at Y2K, which I remember when that was the end, and I almost wonder, you know, maybe would we have been better off if it didn't then or not?

00:07:05.920 --> 00:07:09.069
Uh, but you, you had mentioned microsoft.

00:07:09.108 --> 00:07:12.076
With ai I mean microsoft, ai one.

00:07:12.076 --> 00:07:21.958
To me, artificial intelligence is a very generic term because ai encompasses a wide spectrum of topics.

00:07:21.958 --> 00:07:24.144
You know, we hear the lg's.

00:07:24.144 --> 00:07:29.326
I can't even cover all of the points for it because, you know, a lot of times people just think of the llms.

00:07:29.326 --> 00:07:37.939
You mentioned chat, gpt and recently we've seen some in the news some other local large language models allow processing locally, so it's there.

00:07:37.939 --> 00:08:09.146
So, with with microsoft and ai and the adoption, or where you see the adoption to B2B to adopt, utilize and gain benefit from the use of AI in the organizations or increase some efficiencies, how do you see and position the Microsoft tools to be able to use these AI features and what are some benefits that you see an organization can get from using AI?

00:08:10.670 --> 00:08:29.382
Yeah, sure, especially, that's a very interesting question In the B2B space especially, microsoft is having very good footprint, especially with the Microsoft 365 Office 365 suits and the way I see is especially the users.

00:08:29.382 --> 00:08:35.686
When chart GPT came, that is a aha moment in the artificial intelligence revolution, right?

00:08:35.686 --> 00:08:36.307
They?

00:08:36.307 --> 00:08:43.168
Everybody thought that it is going to take some time, but the user interface for chart GPT is a prompt.

00:08:44.028 --> 00:08:46.494
I looked at the landscape in the enterprise computing.

00:08:46.494 --> 00:08:49.188
Who are having that prompt readily available?

00:08:49.188 --> 00:08:52.568
I see there are broadly three players.

00:08:52.568 --> 00:09:02.530
One is Microsoft Teams and another one is Slack, which Salesforce own, and third one is Zoom right Zoom calls.

00:09:02.477 --> 00:09:03.605
People are used to this video Now they started.

00:09:03.605 --> 00:09:04.385
You know the charting also.

00:09:04.385 --> 00:09:04.573
One is zoom right zoom calls.

00:09:04.573 --> 00:09:05.144
People are used to this video now they are.

00:09:05.144 --> 00:09:05.533
They started.

00:09:05.533 --> 00:09:07.576
You know the charting also.

00:09:07.576 --> 00:09:29.607
So those are the three predominant players for humans to have that kind of interaction from B to C space where chart GPT and Tropic, google, gemini and other players are there to convert that B to C space, that chart prompt experience, into the enterprise experience of the business users.

00:09:29.607 --> 00:09:43.751
I feel like these three predominant companies like Microsoft with Teams, salesforce with Slack and Zoom, are the three players that are going to be really taking this enterprise AI to the next level.

00:09:43.751 --> 00:09:47.248
Those are the three players that are going to be really taking this AI enterprise AI to the next level.

00:09:47.248 --> 00:09:53.888
Those are the user interfaces, because people already have the experience of using prompting the chart.

00:09:55.842 --> 00:10:08.394
I hear the word prompting with an AI and I hear individuals talk about how to become a prompt engineer or prompting tips and tricks for prompting.

00:10:08.394 --> 00:10:13.172
What is prompting and how does someone come about with the prompting?

00:10:13.172 --> 00:10:23.331
And we're talking with large language models and prompting how can those be utilized within the B2B space?

00:10:23.331 --> 00:10:32.149
How does someone understand what prompting is and maybe how to construct a prompt to get the results that they're looking for accurately?

00:10:32.149 --> 00:10:51.589
But I also want to hopefully get into also this new thing that I'm hearing about, which is agents, to where maybe it expands to a little bit more than just prompting or typing for information getting information back, where you have an agent that can possibly do something.

00:10:51.589 --> 00:11:06.816
So it will take some tasks that are possibly repetitive or tasks that can be automated in a sense to allow for someone to have more time and opportunity to do other tasks.

00:11:06.816 --> 00:11:11.447
So how does that all fit within the B2B space?

00:11:11.447 --> 00:11:13.371
How does the prompting work?

00:11:13.371 --> 00:11:18.337
What can you do with the prompting and also then with these agents that are being created?

00:11:18.379 --> 00:11:28.664
I know within Business Central we see a lot of news about Microsoft adding agents and agent previews that are available and talking about that.

00:11:28.664 --> 00:11:30.033
It's not even within Business Central.

00:11:30.033 --> 00:11:31.119
I see the word agent everywhere.

00:11:31.119 --> 00:11:31.981
I think it's going to be.

00:11:31.981 --> 00:11:38.783
I think the word of 2025, if we could talk about it would be agentification or agentizing Agentic.

00:11:39.484 --> 00:11:40.784
I hear that too Agentic.

00:11:40.865 --> 00:11:48.754
Yes, yeah, that's a very good question and very reflective on the introspective question.

00:11:48.754 --> 00:11:49.655
What is prompt?

00:11:49.655 --> 00:11:55.032
Prompt is nothing, but, at least in my words, prompt is nothing but asking a question.

00:11:55.032 --> 00:11:58.330
How do you ask a question is a prompt.

00:11:58.330 --> 00:12:03.293
How do you ask a question to a computer?

00:12:03.293 --> 00:12:06.750
In this case, the AI bot is a prompt.

00:12:06.750 --> 00:12:16.580
The way you ask a question and the way you respond to a question is also a very interesting leadership insight.

00:12:16.580 --> 00:12:22.610
I read a book, or I listened to a book called how Great Leaders Ask Questions.

00:12:23.539 --> 00:12:40.490
So the way we structure the question and what is the strategies that we can use in order to structure your question enables the person to gather more information.

00:12:40.490 --> 00:12:57.129
So this prompt is nothing but the way you ask a question to the bot or AI agent, right, ai on the other side, and the AI computer or AI bot or AI chart, we call it in the Microsoft setup, we call copilot, right, ui.

00:12:57.129 --> 00:13:03.246
So the way we structure the prompt involves different strategies.

00:13:03.246 --> 00:13:05.413
Right, first, we can give the context.

00:13:05.413 --> 00:13:10.884
We say that, hey, what is the news today Is a prompt.

00:13:10.884 --> 00:13:13.187
We can ask that as a prompt.

00:13:13.187 --> 00:13:19.427
Or what is the news today in the United States in the financial sector Is more specific.

00:13:19.427 --> 00:13:26.690
So we are able to structure it and ask a question to get a, you know, intended answer for us.

00:13:26.690 --> 00:13:32.648
So prompt depends on how fine-grained means how specific we are.

00:13:32.648 --> 00:13:39.705
The answer is going to be that much, you know, clear from the ai agents or ai bots.

00:13:39.705 --> 00:13:42.489
So you touch a lot of topics, br.

00:13:42.489 --> 00:13:42.828
Brad.

00:13:42.828 --> 00:13:47.535
So I agree with you this 2025 is going to be the age of agents.

00:13:47.535 --> 00:13:58.049
You know, when we talk about agents, I remember the movie that I watched in 1998, matrix, right, I'm sure everyone remembers about that movie.

00:13:58.049 --> 00:13:59.273
You know the agents.

00:13:59.273 --> 00:14:08.514
So the difference between the way I look at it is the difference between agent is an autonomous thing.

00:14:08.514 --> 00:14:10.847
That's what Salesforce is also calling them.

00:14:10.847 --> 00:14:18.145
And Salesforce came up with the agent force as one of their solutions and they are going full-fledged.

00:14:18.326 --> 00:14:40.825
How Microsoft came with Copilot, satya Nadella, who is the CEO of Microsoft, very clearly articulated that Copilot is the user interface that humans are going to interact with the LLMs or the AI machines, right, and the backend is going to be the agents who are going to do the work.

00:14:40.825 --> 00:14:54.431
If we try to do that in the, you know, correlate that space into the Power Platform, I feel like agents are nothing but Power.

00:14:54.431 --> 00:15:00.623
Automate, right, they are nothing but a Power Automate workflows right.

00:15:00.623 --> 00:15:06.131
Co-pilot, when the user interface, when user prompts or ask a question.

00:15:06.131 --> 00:15:08.842
That goes to the agents, microsoft.

00:15:08.842 --> 00:15:20.461
Interestingly, in one of their documentation they referred agents in three ways One is a responsive agent, another one is a task-based agent and third one is autonomous agent.

00:15:20.461 --> 00:15:32.350
So I feel like, chris, you are know when we talk about agent, which can go and do a task and come back, it is like a power automate flow, right?

00:15:32.350 --> 00:15:39.993
People who are aware of this Microsoft power platform knows what power automate is, which is nothing but a RPA space.

00:15:39.993 --> 00:15:47.360
Uipath is another company that you know they do in the RPA space that they provide the solutions.

00:15:47.360 --> 00:16:00.953
So for our context, agent is nothing but a you know, a software program in the back end that goes and completes a task without giving the information.

00:16:01.115 --> 00:16:02.899
Then what is the difference between?

00:16:02.899 --> 00:16:07.967
Now comes the question what is the difference between power automate and the agent?

00:16:07.967 --> 00:16:08.509
Right?

00:16:08.509 --> 00:16:13.120
Power automate we use to go in the power automate.

00:16:13.120 --> 00:16:21.125
If I want to create a flow, I need to go and drag and drag and drop all the required components.

00:16:21.125 --> 00:16:23.671
What is the trigger, what it needs to do?

00:16:23.792 --> 00:16:25.856
The power automate means send it a email.

00:16:25.899 --> 00:16:42.551
Let us take a simple use case, right, if we want to read an email based on the incoming email, I just want to create an Excel sheet or Word document and send that information back to a team.

00:16:42.551 --> 00:16:56.625
If I take that use case in order, for If I take that use case, in order for us to do that use case right now in the Power Automate, I need to go and create a trigger, say that, hey, incoming email is the trigger to this email box.

00:16:56.625 --> 00:17:04.648
Once we get that email, then do this processing, read the email and create that Excel or Word and send that information to the teams.

00:17:04.648 --> 00:17:07.355
I need to go and create that Excel or Word and send that information to the teams that I need to go and do that.

00:17:07.355 --> 00:17:12.928
But Microsoft, now recently they created a copilot for Power Automate.

00:17:12.928 --> 00:17:16.945
Now I can go to the copilot and say that, hey, create this workflow.

00:17:16.945 --> 00:17:29.126
So this workflow of reading the, you know, anticipating for the email and reading the email and creating a Word document or Excel and sending it to them.

00:17:29.126 --> 00:17:31.888
So, stepping back really quick Sai.

00:17:32.220 --> 00:17:34.968
You mentioned Copilot, basically more of a.

00:17:34.968 --> 00:17:40.380
The way I look at it sounds like to me Copilot is more of a translator.

00:17:40.380 --> 00:17:52.692
You ask a prompt of what you want based on what's available for you within your maybe tenant, then it chooses the correct agent to respond.

00:17:52.692 --> 00:17:54.846
So it's almost like a translator.

00:17:54.846 --> 00:18:07.894
Right for that prompt Because, as you know, in the B2B space, when you're creating or you're interacting with Copilot within your organization, it should only respond based upon what's available to it.

00:18:09.375 --> 00:18:09.576
Right.

00:18:09.880 --> 00:18:11.586
So you are right, chris.

00:18:11.586 --> 00:18:15.310
So the Copilot is like you said.

00:18:15.310 --> 00:18:17.027
It's basically an interface.

00:18:17.027 --> 00:18:21.078
It does some operation, it manages the agents.

00:18:21.078 --> 00:18:43.925
You can say that it's an orchestration piece where it takes the information from the user and, based on the available agents, it will direct the agents, orchestrate the agents to go sequence of tasks and come back and provide the answer to the user, to the human or to the user in the NLP, natural language processing.

00:18:43.925 --> 00:18:51.148
So now the way we interact with the co-pilots or AI agents is completely changed.

00:18:51.148 --> 00:18:55.848
From the mouse, we take that and click that different buttons to get the information.

00:18:55.848 --> 00:19:08.115
Now we are using natural language processing to talk to, like how we are able to talk to other human, like how we are discussing, we are able to just enter the information to the co-pilot.

00:19:08.700 --> 00:19:12.671
Microsoft is having their own lab called co-pilot.

00:19:12.671 --> 00:19:14.226
You know Microsoft Labs.

00:19:14.226 --> 00:19:17.769
They are experimenting with voice also.

00:19:17.769 --> 00:19:22.702
So, like how we are discussing, they have a co-pilot voice, the co-pilot voice.

00:19:22.702 --> 00:19:27.523
We can just enable the voice and we can say that, hey, this is the task that we want to do.

00:19:27.523 --> 00:19:36.967
Then you know, it can go ahead and create the agents and orchestrate the agents and come back and with the answer.

00:19:37.539 --> 00:19:48.583
So in one of the recent interviews also, I think, satya Nadella, ceo of Microsoft, he told that SaaS kind of you know, in the future SaaS may be.

00:19:48.583 --> 00:19:50.609
What is SaaS applications?

00:19:50.609 --> 00:19:53.086
Saas is software as a service applications.

00:19:53.086 --> 00:19:56.686
Right, they are basically a CRUD.

00:19:56.686 --> 00:20:01.406
Applications means they are having a database On the top of the database.

00:20:01.406 --> 00:20:07.119
The user interface provides the user to interact to perform the CRUD operations.

00:20:07.279 --> 00:20:20.250
If we take CRM right, crm is having a sales module in that there are certain database tables which are in the Power Platform called Dataverse.

00:20:20.250 --> 00:20:34.723
Sales module provides the user interface for the users to go ahead and create codes, purchase orders, leads and opportunities, all those things.

00:20:34.723 --> 00:20:40.277
In the future, what is going to happen is people are expecting that may be sooner, maybe within the next few years.

00:20:40.277 --> 00:20:55.023
Instead of user going to the sales application and entering the information, people will go to the prompt copilot, sales copilot and they say that, hey, this is a new lead that I got, this is the you know.

00:20:55.023 --> 00:20:55.727
Take the picture.

00:20:55.727 --> 00:21:03.011
Say that, hey, create a lead information in the sales of Dynamics 365.

00:21:03.011 --> 00:21:05.307
It should be able to create that information.

00:21:05.307 --> 00:21:14.251
So the user experience itself may be, you know, completely changing the way users interact with these enterprise applications.

00:21:14.251 --> 00:21:16.366
Maybe really changing.

00:21:17.382 --> 00:21:34.859
That could take me down a completely separate path because Chris and I recently spoke about that as well as far as how we interact with data, how we retrieve data and having the ability to use natural language to interface with that.

00:21:34.859 --> 00:21:42.814
But I'm still trying to go way back to the beginning of prompting to get information out.

00:21:42.814 --> 00:22:17.166
How do we come up with a and how do we learn and how do we know to come up with the proper prompt either for to go back to the points that you had mentioned, either it's data retrieval or language and learning I type, you know, create me a picture or ask some information based upon the data that the model has been trained on or in the construct of what you and Chris had mentioned, with the Power Platform to utilize Copilot Studio in a sense, which I want to get into to create these tools for us, basically our own agents.

00:22:17.166 --> 00:22:26.880
But where I get confused is we mentioned task-based agents how is their variability in tasks?

00:22:26.880 --> 00:22:43.590
Because I still say, something that I tried to do, I wish I could do, is even something as simple as scheduling, taking my emails, taking a look at my calendars to be able to automatically reply, like even with the podcast, for example, we do a lot of scheduling of the guests, such as yourself with the podcast, with the.

00:22:43.611 --> 00:22:50.252
We do a lot of scheduling of the guests, such as yourself with the podcast, with the pre-podcast planning calls.

00:22:50.252 --> 00:23:08.111
Chris, you have to fix that Pre-podcast planning calls to the actual schedule of the recording taking a look at calendars, taking a look at time zones to offer and suggest times that best fit based upon availability time zone and such times that best fit based upon availability, time zone and such.

00:23:08.111 --> 00:23:10.794
There's a lot to that, and is that something that could be done and how could you do that?

00:23:10.794 --> 00:23:12.997
Is that multiple agents within Power Automate?

00:23:12.997 --> 00:23:19.000
No-transcript.

00:23:42.726 --> 00:23:45.528
Right, so now you give a it's all within that space.

00:23:46.250 --> 00:23:50.020
So, utilizing that, how could I do that?

00:23:50.020 --> 00:23:54.122
I hear a lot about Copilot and I hear a lot of things that we have agents that can do anything.

00:23:54.122 --> 00:23:56.461
I'm just trying to see a practical use and example of it.

00:23:57.749 --> 00:24:10.380
Yeah sure, so let's take that use case that you mentioned about the podcast right For us to create this Microsoft AI agents or Microsoft co-pilots broadly.

00:24:10.380 --> 00:24:13.765
There are two ways that we can do it right now in the Microsoft platform.

00:24:13.765 --> 00:24:15.732
One is using the.

00:24:15.732 --> 00:24:27.720
Microsoft came up with a co-pilot studio that is part of the Power Platform that provides the tools and knowledge bases and inbuilt agents also that enable the user.

00:24:27.720 --> 00:24:34.641
That's a low-code, no-code platform Copilot Studio, where the users can go ahead and create the AI agents.

00:24:34.641 --> 00:24:41.080
And another way to do that in the Microsoft platform is Azure AI Foundry.

00:24:41.080 --> 00:24:44.084
Microsoft just recently launched Azure AI Foundry.

00:24:44.084 --> 00:24:47.798
Microsoft just recently launched Azure AI Foundry, which is based on them.

00:24:47.798 --> 00:24:50.214
We can go ahead.

00:24:50.214 --> 00:25:15.402
We can use the Azure AI Foundry and create the agents using different LLMs that are available, such as we can use OpenAI Microsoft is having 49% stake in the OpenAI, so they create exclusive access to the OpenAI models or we can use Anthropic models, or we can use LAMA, which is Meta's open source AI models.

00:25:15.402 --> 00:25:17.577
So broadly, we can do it in two ways.

00:25:17.577 --> 00:25:23.980
One is Copilot Studio Microsoft Copilot Studio or Microsoft Azure AI Foundry.

00:25:23.980 --> 00:25:26.173
For our conversation.

00:25:26.513 --> 00:25:28.138
I have good experience in.

00:25:28.138 --> 00:25:37.134
You know, I created a couple of agents in the using Copilot Studio so we can do the use case that you mentioned using the Copilot Studio.

00:25:37.134 --> 00:25:41.936
Copilot Studio is a very easy way for us to create the agents.

00:25:41.936 --> 00:25:45.259
Previously Microsoft used to call as co-pilots.

00:25:45.259 --> 00:25:47.898
They renamed it a few months back to agents.

00:25:47.898 --> 00:25:53.380
So to create any agent we need broadly two or three things.

00:25:53.380 --> 00:26:02.403
First one is what is a knowledge base right, based on what the agent need to create the information.

00:26:02.403 --> 00:26:03.816
Second one is tasks.

00:26:03.816 --> 00:26:09.420
These tasks are nothing but the tasks that just you outlined to create this podcast.

00:26:09.420 --> 00:26:12.634
We need to look at, you know first, evaluate this.

00:26:12.634 --> 00:26:18.355
You know participants, send them an email and have the review session so we can create.

00:26:18.355 --> 00:26:23.574
We need to break down into different tasks and that task.

00:26:24.135 --> 00:26:27.657
Microsoft is very good, especially in the Coopilot Studio.

00:26:27.657 --> 00:26:29.396
They came up with a lot of connectors.

00:26:29.396 --> 00:26:47.104
We can connect with the Outlook, which is a native thing, so we can easily connect to the Microsoft Outlook and create a task and send an email If we want to talk to any other databases or like Riverside if we are trying to use, task and send an email If we want to talk to any other databases or like Riverside, if we are trying to use.

00:26:47.104 --> 00:26:52.718
We will go ahead and see whether the Co-Pilot Studio is having any Riverside connectors.

00:26:52.718 --> 00:26:56.759
If not, users can create that custom connectors.

00:26:56.759 --> 00:27:01.354
So Microsoft enabled all these features for the low code.

00:27:01.354 --> 00:27:05.240
No code developers to create this kind of connectors to create the agent.

00:27:05.240 --> 00:27:07.718
So the agent can be created.

00:27:07.718 --> 00:27:14.778
First, this podcasting use case we can break down into tasks and each task we can go ahead and create.

00:27:14.778 --> 00:27:19.622
It is similar to creating a workflow in the Power Automate.

00:27:19.622 --> 00:27:21.876
You're muted, by the way, brad.

00:27:21.876 --> 00:27:22.873
You are muted.

00:27:22.932 --> 00:27:30.258
Brad, don't tell anybody, because I was getting excited and I had to mute myself because I wanted to hold back.

00:27:30.258 --> 00:27:35.836
So I can create an agent that will send an email to Sai.

00:27:35.836 --> 00:27:41.755
Sai, we'd like to speak to you on the podcast, are you okay?

00:27:41.755 --> 00:27:42.478
I'm simplifying.

00:27:42.478 --> 00:27:52.991
It may be one use case In the other cases where individuals contact us and say they would like to speak with us about a topic which we enjoy getting those emails as well.

00:27:52.991 --> 00:27:58.142
So I can say email Sai, ask him to be on the podcast.

00:27:58.142 --> 00:27:59.569
You'll reply yes.

00:27:59.569 --> 00:28:06.256
So the agent can reply to that email, knowing that the original email was sent out as a request to the podcast.

00:28:06.256 --> 00:28:17.394
Response is whatever verbiage is yes, and then the agent can respond and say okay, let's do this, let's set up a planning call.

00:28:17.394 --> 00:28:19.115
Is this time good for you?

00:28:19.115 --> 00:28:20.788
Or here's these times that are good for you?

00:28:20.788 --> 00:28:23.531
Based on rules or based on Based?

00:28:23.612 --> 00:28:24.414
on our calendars.

00:28:24.414 --> 00:28:26.623
If we we could do this, this would be amazing.

00:28:27.164 --> 00:28:28.609
So I can send out the email.

00:28:28.609 --> 00:28:30.336
The response would come back.

00:28:30.336 --> 00:28:36.013
It would automatically send dates based upon a calendar for availability.

00:28:36.013 --> 00:28:46.817
The participant would be able to then respond with this works best for me and then it would automatically schedule and put in the text that I like to use.

00:28:46.817 --> 00:28:52.696
And all that and the link, yeah, because the studio link is yeah, that's, that's correct, right.

00:28:52.777 --> 00:28:53.699
Yeah, that's correct, chris.

00:28:53.699 --> 00:28:57.952
So what you know, now we are getting into very we need to do this afterwards.

00:28:57.992 --> 00:29:00.355
If we can really do this, I would like to schedule time.

00:29:00.355 --> 00:29:06.303
You can certainly set it up is this something you can do and then send to me?

00:29:06.303 --> 00:29:09.207
Yes, I think we can try that.

00:29:09.207 --> 00:29:31.278
We'll do a follow-up because I want to see this in action, because that is such a good experience and good use of the tools, as well as saving time yeah, not only that, you that you know you will have, you know this Chris and Brad will have their own agents right To send the email to schedule this.

00:29:31.549 --> 00:29:32.634
I will have my own agent.

00:29:32.634 --> 00:29:37.121
So as soon as I, that is who could do my email right.

00:29:37.121 --> 00:29:44.583
So as soon as I see any request or any information if I want, my agent will respond to your agent.

00:29:44.583 --> 00:30:03.701
I can create a small agent, say that, hey, if I am going for this podcast or speaking once I get an email, I create a small agent and say that, hey, just respond back with my email ID, with my calendar availability next couple of weeks to your agent.

00:30:03.701 --> 00:30:08.642
So this is going to be really agent orchestration.

00:30:08.741 --> 00:30:15.701
Right At your end you will have a couple of agents which will be triggering the email or receiving the email.

00:30:15.701 --> 00:30:20.959
I can just next time I will send an email say that, hey, I'm interested in your podcast.

00:30:20.959 --> 00:30:27.400
Or I will just ask co-pilot say that, hey, next couple of weeks I am interested to.

00:30:27.400 --> 00:30:36.884
I have this time I want to really talk and get to know more about what is happening in the dynamics with Brad and Chris.

00:30:36.884 --> 00:30:53.461
So I will just ask the co-pilot and the co-pilot goes and talks to my agent and send an email to you guys and your agent can pick it up and look at your availability and schedule and confirm something back to my agent.

00:30:53.670 --> 00:30:57.790
There's actually two places where I can see this working Right now.

00:30:57.790 --> 00:31:03.675
Our website has a place for you to be a guest, right, you fill out your information and stuff and we get an email.

00:31:03.675 --> 00:31:13.258
So we could use a Power Automate to collect that information and, based on that information, then we can use Copilot to act upon that.

00:31:13.258 --> 00:31:22.500
Where it looks at our calendars, make sure that it answered you know we got all the information it needs and then it can, you know, schedule that for us.

00:31:22.500 --> 00:31:29.798
And number two we can even put copilot on our website, probably, and sai can interact with it.

00:31:29.798 --> 00:31:45.733
It's going to ask all the questions that the form would have asked anyway, collect that information based on size responses, notifies us and looks at our calendars and let us know like hey, si's interested in this and then schedule it out and we just show up and have a conversation.

00:31:46.455 --> 00:31:51.511
I I like the use case because it's to go back to where I started with.

00:31:51.511 --> 00:31:52.492
This is.

00:31:52.492 --> 00:32:01.896
We hear all this ai agent, we hear prompting, we hear models, we hear all this, but I hear it.

00:32:01.896 --> 00:32:04.602
Well, you can do anything or you can do specific tasks.

00:32:04.602 --> 00:32:11.343
Now I'm trying to just put my head around something as simple as scheduling with somebody to be on a podcast.

00:32:11.343 --> 00:32:13.292
Somebody could do this on their own.

00:32:13.313 --> 00:32:24.845
For something else, even lawn maintenance, if somebody has a landscaping or a lawn maintenance company or even electrical services, you know any type of scheduling that you need to go through.

00:32:24.845 --> 00:32:26.894
We usually have an individual going back and forth.

00:32:26.894 --> 00:32:32.213
I know often I'll use what they used to call fine time or whatever.

00:32:32.213 --> 00:32:36.777
That is where you can say, okay, here's several dates and times, pick the ones that work the best and then from that.

00:32:36.777 --> 00:32:41.711
But that's, to me, is not such an elegant experience sometimes.

00:32:41.711 --> 00:32:46.561
So I would like the more personal interaction of I'm sending an email to Cy.

00:32:46.561 --> 00:32:47.544
Step one, chris.

00:32:47.544 --> 00:32:50.058
I like the idea of also taking ingesting on the inside.

00:32:50.058 --> 00:32:54.317
Then all of a sudden stuff shows up on the calendar and then we just just do it.

00:32:54.317 --> 00:32:55.490
I think that's also great.

00:32:56.092 --> 00:33:12.255
I think you can even prompt it where like, do not schedule anything just within the space because maybe you're already busy, and then look at, you know if there's already existing one and you can specify it's the only schedule between these days or between these times.

00:33:12.576 --> 00:33:13.318
Oh, we have to do this.

00:33:13.318 --> 00:33:20.163
Okay, so we can set it up to where we have an agent in Power Automate that will send an email.

00:33:20.163 --> 00:33:29.941
It will send an email based on me just saying send an email to Sai to be on the podcast, right, yeah.

00:33:30.000 --> 00:33:30.643
Just like that.

00:33:30.990 --> 00:33:40.420
We can easily do that, yes, and then we'll send an email to Sai with a template with the information that's pertinent to Sai so that he understands what the podcast is.

00:33:40.420 --> 00:33:41.855
Then you'll reply.

00:33:41.855 --> 00:33:53.813
The agent will read the reply and we can say go look at this calendar and this calendar and propose some times based on size, time zone.

00:33:53.813 --> 00:33:56.641
I like this.

00:33:56.641 --> 00:33:58.175
That is important.

00:33:58.175 --> 00:34:01.119
I'm trying to think of the variability here because these are the scheduling challenges.

00:34:01.119 --> 00:34:02.906
We recorded all hours of the day to make accommodations are the scheduling challenges.

00:34:02.906 --> 00:34:07.159
You know we recorded all hours of the day to make accommodations for everyone's schedule.

00:34:07.159 --> 00:34:10.596
That that's going to so how much how much time you spend, brad.

00:34:10.695 --> 00:34:13.222
I think uh looks like it's a lot of efforts for you.

00:34:13.222 --> 00:34:15.434
You know, to host this kind of podcast right.

00:34:15.434 --> 00:34:17.460
Look at the look at the just calendar.

00:34:17.480 --> 00:34:27.786
Simple task of calendar it's looking at calendars, going back, making sure that it's a lot of appropriate for the guests because, as we say, it's it's anybody who's been on.

00:34:27.865 --> 00:34:31.797
We, you know, in the planning call, we talk about that when they're at their best, when it fits them.

00:34:31.797 --> 00:34:44.690
We try to work around trips if somebody has conferences to go to, for example, or if there are holidays and and sometimes individuals don't mind, you know we'll do a recording on a holiday or something.

00:34:44.690 --> 00:34:54.780
So there are some variables in there that, based upon, we may need to find out which times work best for you, but it is a lot to juggle schedules for many calendars.

00:34:55.090 --> 00:34:57.039
So how do, how do you guys do?

00:34:57.039 --> 00:34:59.028
Do you guys outsource that piece of?

00:34:59.048 --> 00:34:59.411
work we do.

00:34:59.512 --> 00:35:09.731
outsource it to me piece of work to me.

00:35:09.751 --> 00:35:10.956
Yeah, man it's a lot of work, right then a lot of energy, guys.

00:35:10.976 --> 00:35:11.978
The scheduling is a lot of work.

00:35:11.978 --> 00:35:24.782
The scheduling is uh, I'd want to say it's a full-time job, but to try to do proper scheduling and I like to do things properly as well as you see, when you go through the experience, to make sure that everybody has an enjoyable experience as they go through this with a little bit of a personal touch as well.

00:35:24.782 --> 00:35:31.543
But, it does say take some time because we have several calls per week that we do on top of everything else.

00:35:32.590 --> 00:35:51.260
You know, it'd be fascinating though, because once we get all this solved right, it's to have a system, or maybe co-pilot or an agent, where he takes all of our files and says I want you to edit this for our podcast video and it just does it for us.

00:35:51.541 --> 00:36:01.630
Right now it's all manual man, we do it man we do it well, chris, does the the post production you know we have a process and it works well.

00:36:01.630 --> 00:36:09.105
Because of timing, I'll do a lot of the scheduling, interfacing with the guests that come on that we are extremely appreciative of everybody that spends the time with us.

00:36:09.105 --> 00:36:10.496
Time is extremely valuable.

00:36:10.496 --> 00:36:15.141
I know to me personally and Chris and I talk about it because it's what we have.

00:36:15.141 --> 00:36:16.856
Once you use it, you don't get it back.

00:36:16.856 --> 00:36:18.054
You don't get a redo of a minute.

00:36:19.271 --> 00:36:20.496
You are a wise man, Brad.

00:36:20.496 --> 00:36:22.934
People realize that very late.

00:36:22.934 --> 00:36:24.338
I am realizing now.

00:36:24.338 --> 00:36:25.481
I think time is the real.

00:36:36.750 --> 00:36:36.849
It is.

00:36:36.849 --> 00:36:37.110
It takes you.

00:36:37.110 --> 00:36:40.077
You do have to get to a certain point in life, I think, where just maturation in life, that this is something I wish.

00:36:40.077 --> 00:36:41.480
Everyone always asks me what would you tell your younger self?

00:36:41.480 --> 00:36:45.827
That's one of the things I would tell myself is one listen to those that have gone through a lot because their experiences.

00:36:45.827 --> 00:36:57.637
I'm not saying you have to listen to the experience as far as following, but sometimes just listening objectively to somebody's experiences and maybe learn from them instead of thinking, ah, they don't understand, I can do this, I can do that.

00:36:58.001 --> 00:37:01.639
And also time, you know where you spend your time and what you spend your time on.

00:37:01.639 --> 00:37:05.793
You cannot value anything more than that.

00:37:05.793 --> 00:37:09.753
But to go back to what we talked about, so I do a lot of the scheduling, chris does the post-production.

00:37:09.753 --> 00:37:17.896
Hopefully at some point we can incorporate some ai into it, um, which thankfully riverside has added quite a bit that we can do some stuff.

00:37:17.896 --> 00:37:20.231
But uh, chris, I don't want to jump the gun yet.

00:37:20.231 --> 00:37:26.077
I'm still trying to go back to the scheduling to save myself some time and we'll set this up and I won't even tell Chris and I'm like I have to go through all this scheduling.

00:37:26.530 --> 00:37:43.684
I think, like you said, you put it well perfectly about the time spent, where I think co-pilot AI in general is really, really important for the future, because at that point, as we use it more, we don't trade time for money.

00:37:43.684 --> 00:38:06.074
We get to a point where I want to trade time for experiences, right, and and that's going to be, um, that's going to be important in the future for it, for my, for my view, because, as like you said, sai, you know, you want to make sure you're you're utilizing your time in the right places and not trading it for money every single time.

00:38:07.097 --> 00:38:07.599
Life is short.

00:38:10.153 --> 00:38:11.076
It really is.

00:38:11.076 --> 00:38:12.097
It's um.

00:38:12.097 --> 00:38:13.842
You know, you go through phases.

00:38:14.170 --> 00:38:17.139
I thought we were now getting philosophical and I don't mean to digress, maybe I'll just stop.

00:38:17.590 --> 00:38:21.001
I could get on the philosophical road forever but um time.

00:38:21.001 --> 00:38:22.418
So we'll go back to the efficiency of time.

00:38:22.418 --> 00:38:25.231
So we can set up an agent to send an email.

00:38:25.231 --> 00:38:33.898
We can set up an agent to read an email and then, based on the contents of the email, work on scheduling the podcast.

00:38:33.898 --> 00:38:36.206
I have to see this work.

00:38:36.887 --> 00:38:39.132
Yeah, that's a good use case, at least in 2025.

00:38:39.132 --> 00:38:42.860
Maybe next, 2026, 2027, we just need to talk to that.

00:38:42.860 --> 00:38:48.117
You know co-pilot or the UI, the prompt say that, hey, this is what we want to do.

00:38:48.117 --> 00:38:49.393
It has to.

00:38:49.393 --> 00:38:51.099
It may be going and doing all this.

00:38:51.099 --> 00:38:54.980
You know scheduling and creating all this for us.

00:38:54.980 --> 00:39:00.251
Maybe right now we need to enter the prompt on the keyboard In the future.

00:39:00.251 --> 00:39:01.657
You know there are NVIDIA.

00:39:01.657 --> 00:39:08.679
All these companies are investing heavily on voice-based, so we just talk to them.

00:39:08.679 --> 00:39:09.853
Hey, do you guys?

00:39:09.853 --> 00:39:14.922
I have a question Do you guys recently OpenAI released Operator?

00:39:14.922 --> 00:39:17.125
Did you guys look at that Operator?

00:39:17.688 --> 00:39:17.829
demo.

00:39:17.829 --> 00:39:27.001
I watched the demo of again you need to have from what I read, you need to have OpenAI Plus or whatever that means, Right, $200.

00:39:27.001 --> 00:39:31.659
The expensive plan, which I understand, but that was impressive as well.

00:39:31.659 --> 00:39:35.159
It would go to a website and interact with the website.

00:39:35.159 --> 00:39:40.760
I saw it do the scheduling, I saw it do reservations and a number of other things.

00:39:40.760 --> 00:39:41.876
That is impressive.

00:39:41.876 --> 00:39:42.320
That's interesting.

00:39:42.340 --> 00:39:42.862
That is impressive.

00:39:42.862 --> 00:39:43.889
Yeah, interesting, that is impressive.

00:39:43.889 --> 00:39:45.615
Yeah, I have a question for both of you.

00:39:45.615 --> 00:39:49.840
Given you guys have so much of experience in the IT side, right?

00:39:49.840 --> 00:39:56.820
So, technology-wise, in the leadership roles and all those, how do you see this enterprise AI evolving?

00:39:56.820 --> 00:40:01.889
So I shared my thoughts, right, and going back to that mainframe era.

00:40:01.889 --> 00:40:03.097
I know a lot of companies like banks, insurance companies.

00:40:03.097 --> 00:40:03.902
Going back to that mainframe era.

00:40:03.902 --> 00:40:06.672
I know a lot of companies like banks, insurance companies.

00:40:06.672 --> 00:40:12.514
They are still on mainframe, given the architectures you know in the enterprise computing.

00:40:12.514 --> 00:40:20.237
You know it is very difficult to move to the latest and greatest, given SLAs and you know lawsuits and all those things right.

00:40:20.697 --> 00:40:30.298
So on the Java side, I worked in Sun Microsystems in India for almost three years, the guy who invented Java also I was able to.

00:40:30.298 --> 00:40:40.429
I was part of one leadership committee way back, so I was able to meet that guy, james Gosling, who invented Java and it's a very interesting experience.

00:40:40.429 --> 00:40:54.005
But when I put that also in the context, java, the latest version is Java 20 plus right Now, oracle, bot, sun, they have 20 plus, but in production they will have old versions of Java.

00:40:54.005 --> 00:41:01.219
Still, I know Java 8, which was released way back in 2006-2007, their enterprises are using it.

00:41:01.219 --> 00:41:14.099
If we put mainframes and Java versions in the enterprise penetration right In the production environments, people will be doing a lot of testing and POCs and sample projects.

00:41:14.099 --> 00:41:29.025
But for enterprise applications or CRM kind of applications, how do you guys see this AI going to, the transition or adoption of this AI in this ERP and CRM space?

00:41:29.731 --> 00:41:32.501
That is a challenging question.

00:41:32.501 --> 00:41:35.759
In my opinion, you brought up some key points.

00:41:35.759 --> 00:41:40.702
It's where is an organization in its journey?

00:41:40.702 --> 00:41:43.753
Which systems do you have?

00:41:43.753 --> 00:41:48.684
Also, which systems and which tools are available?

00:41:48.684 --> 00:41:50.434
I think you need to have a combination.

00:41:50.434 --> 00:41:52.420
Should you always have the latest and greatest?

00:41:52.420 --> 00:41:52.820
I don't know.

00:41:52.820 --> 00:41:56.679
I think, in my opinion, sometimes there's risk and you need to evaluate what you use.

00:41:56.679 --> 00:42:03.425
I think for those more mature organizations, there may be pieces that you can plug in.

00:42:03.425 --> 00:42:20.195
So if you think of even going through to bring it back to a point, to a business, central implementation or implementation of an ERP application, it's a matter of architecting a solution with the right pieces and putting those right pieces together to get the desired results.

00:42:20.916 --> 00:42:36.175
But individuals also need to almost change their way of thinking at some points to ask questions outside of the predefined constraints, Because a lot of times people make decisions based on the past.

00:42:36.936 --> 00:42:41.824
I know, running through this woods I ran into a bear.

00:42:41.949 --> 00:42:44.143
So now I'm going to make sure every time I come through this point I'm going to ran into a bear.

00:42:44.143 --> 00:42:45.936
So now I'm going to make sure every time I come through this point I'm going to run into a bear.

00:42:45.936 --> 00:42:53.235
But now I need to make sure, if I go down this path, I do all this stuff but in reality, that bear may never be there again.

00:42:53.235 --> 00:43:20.516
So you have to take off the constraints of the limitations you had based upon the past and not think that you need to do something in totality to where you have such a radical shift, but maybe compartmentalize the pieces to incorporate those changes Again the efficiency that you can get in the AI to where you can increase the adoption within your organization and ensure that it's going to give you the desired results as well.

00:43:20.516 --> 00:43:36.934
It's a lot there of what I am saying, but I think organizations need to evaluate where they can get the benefit of using a tool, making sure they use the right tool and don't use the tool just to use the tool.

00:43:38.675 --> 00:43:41.378
Yeah, that's a good point, Brad.

00:43:41.378 --> 00:43:48.204
From my perspective, there are two different paths, because there's still some human element.

00:43:48.204 --> 00:43:56.652
As you know, when someone uses a technology on their personal life, they typically bring that to work, expecting to do the same thing.

00:43:56.652 --> 00:44:04.282
There's still a lot of education that still needs to happen when it comes to AI.

00:44:04.282 --> 00:44:18.601
I've had a lot of conversation, a lot of people, even day-to-day people, that talks about AI and they think it's a one one solution fits all and, as we know, in our space, that's not always the case.

00:44:18.601 --> 00:44:36.380
That's not the case at all, because now we're talking about agentic or agents that are used in the back end to do some specific task or give you results or just have a plain conversation of when you're having that utilizing ai.

00:44:36.380 --> 00:45:02.487
So I I still think we have a little bit of time that we need to educate everybody that there are differences between the two, because you talked about Gemini right as another LLM, but then you also have Copilot and so from a public perspective, it's just an AI to have a good natural language conversation, but in our space, in the enterprise space, that's not the case.

00:45:18.340 --> 00:45:21.865
But in our space, in the enterprise space, that's not.

00:45:21.865 --> 00:45:43.139
We're going to have systems talk to each other and what comes to that means there are going to be things or, unfortunately, positions that are going to be replaced because of those co-pilots or because of those agents, and I think, from a business standpoint, that's going to be a place where you need to plan.

00:45:43.139 --> 00:45:44.054
You know what does that?

00:45:44.054 --> 00:45:46.760
What does that mean for your business?

00:45:46.760 --> 00:46:00.221
Um, that means more time for you, maybe more time to be more creative, but even then, creativity could be replaced as well, it's like ai, it's, it's one of those never-ending cycles, but we also have to.

00:46:00.402 --> 00:46:01.563
It goes back with time.

00:46:01.563 --> 00:46:09.797
It's sometimes you have to worry about what's in front of you versus so far into the future, because you don't know and nobody can predict what will happen.

00:46:09.797 --> 00:46:26.822
Uh, the any businesses have evolved since back in the early days when they use ledgers with ink pen, inkwells and pens, right quill pens, and now then you had the ballpoint pen and the pencil, and then you went to computers with spreadsheets.

00:46:26.822 --> 00:46:44.568
So there's always been an evolution in of efficiencies and gaining of those efficiencies and then just a reallocation of talent to do those tasks that haven't been to the point where they have been as made, optimal, I guess you could say, or added the efficiencies.

00:46:44.568 --> 00:46:45.329
So it's.

00:46:46.655 --> 00:46:52.108
Yeah, I think both your points are very valid and very interesting perspectives.

00:46:52.108 --> 00:46:55.684
The way I look at is the ROI right.

00:46:55.684 --> 00:47:00.007
As a business, they will see what is the ROI on their investments.

00:47:00.007 --> 00:47:13.228
So, especially with the teams summarization now when I am using teams in my workplace after the meeting previously I used to take the notes and all that information.

00:47:13.228 --> 00:47:17.163
There will be someone who takes the notes and share that after the meeting.

00:47:17.163 --> 00:47:24.427
Now, with the team scope, we can get a summary of the meeting pretty quickly.

00:47:24.427 --> 00:47:27.672
Perfect, that is a very good ROI.

00:47:27.672 --> 00:47:29.239
For me, that is a great use case.

00:47:29.335 --> 00:47:33.746
I don't mean to cut you off, but with Teams turning on the transcription to record it.

00:47:34.036 --> 00:47:40.880
some individuals get a little nervous, thinking I'm going to do the video recording or someone has the recording, but I agree with you.

00:47:40.880 --> 00:47:44.059
Recording, or someone has the recording, but I agree with you.

00:47:44.059 --> 00:47:58.304
Just something as simple as doing the transcription of the voices gives you the benefit in the meetings that those participants can pay attention to what's being discussed instead of worrying about the notes that they have to take, because you cannot do both tasks at once.

00:47:58.304 --> 00:48:04.545
If you're spending time trying to write the proper notes, you're not listening.

00:48:04.545 --> 00:48:06.541
I don't care what anybody says.

00:48:06.541 --> 00:48:12.588
They think that they can multitask listen and basically listen and talk at the same time.

00:48:12.655 --> 00:48:18.043
So I just want to bring up something as simple as that is a huge gain.

00:48:18.043 --> 00:48:23.893
And you get actual summaries, and I don't know why.

00:48:23.893 --> 00:48:33.963
I wish I could have it set where it says okay, record any call that I jump into for the transcription so I don't forget, because there have been times like, ah, I wish I had this on and I forgot.

00:48:34.074 --> 00:48:40.135
Do you remember, on meetings like that, where you have someone's responsibility and that's all they did was note-taking?

00:48:40.135 --> 00:48:50.277
If you guys recall back in the day, where you sit in a Conference you have somebody sit in the back corner, that's all they did was take notes, right, that's my point.

00:48:50.277 --> 00:48:55.800
Right, like there's gonna be a shift where that role is no longer needed in.

00:48:55.800 --> 00:48:59.065
It's a lot more accurate for note-taking, right, and and so it's.

00:48:59.065 --> 00:49:07.010
It's always always listening when a human may be distracted, and so it's always always listening where a human may be distracted.

00:49:07.010 --> 00:49:08.012
And they forgot a specific note.

00:49:08.012 --> 00:49:13.603
Now, with Copilot, to be able to do that for you and summarize and even ask, like, what was the action items out of this meeting?

00:49:13.603 --> 00:49:31.148
Because you want it to be a productive meeting, it's going to tell you, versus having to like ask that person, say, hey, can you type that all up, and then by the time you get those notes, it's the end of the day where Copilot can give you that information right after that meeting.

00:49:32.315 --> 00:49:41.748
Yes, I think that you are right that jobs and all those they need to be retrained or they will be going into a different place in this era.

00:49:41.748 --> 00:49:45.184
That is one of the reasons in the Copilot Studio.

00:49:45.184 --> 00:49:49.766
Microsoft enables us as soon as we create the agent.

00:49:49.766 --> 00:49:53.664
It allows us to publish to the teams first Teams are.

00:49:53.664 --> 00:49:55.759
It gives a different channels where we can publish.

00:49:55.759 --> 00:50:00.266
It Looks like the teams is one of the very efficient way to interact.

00:50:00.266 --> 00:50:04.596
And another use case also recently I was working with one of the very efficient way to interact.

00:50:04.579 --> 00:50:06.592
And another use case also recently you know I was working with one of the clients.

00:50:06.592 --> 00:50:09.603
They had a lot of knowledge base.

00:50:09.603 --> 00:50:15.565
They went with a big implementation right of their ERP but that was not successful.

00:50:15.565 --> 00:50:19.284
So they were trying to evaluate and we got RFP.

00:50:19.284 --> 00:50:27.079
I was trying to me and my team members were trying to look at their business models or business rules.

00:50:27.079 --> 00:50:28.762
What exactly is their business?

00:50:28.762 --> 00:50:34.143
So I cannot share the client details, but it is in the healthcare sector.

00:50:34.143 --> 00:50:35.880
There is a lot of information.

00:50:35.880 --> 00:50:39.300
They went for the ERP implementation for almost one, two years.

00:50:39.300 --> 00:50:40.204
It was not successful.

00:50:40.204 --> 00:50:41.619
We got a RFP.

00:50:42.255 --> 00:50:54.025
So generally the process was go through all the documentation to understand their business, specific business, because you know their business process and what are the challenges that they faced.

00:50:54.025 --> 00:51:04.083
So what we did was we took their Word documents, their PDFs, their audio, their video recordings and downloaded the transcripts.

00:51:04.083 --> 00:51:17.788
We combined all those things and we uploaded to the SharePoint and created a co-pilot on top of it and started asking questions to the co-pilot Say that hey, what is the business process?

00:51:17.788 --> 00:51:20.581
Can you explain in broad I know high level 10 steps.00:51:20.581 --> 00:51:25.643


Surprisingly, it was able to give us at least six, seven steps correctly.00:51:26.114 --> 00:51:48.001


So we started fine tuning the prompts and adjusting, given what happens is this copilot, once it goes to the knowledge source, in our case the SharePoint documentation, the backend, it indexes and it creates a structure so that the co-pilot can efficiently go and read and give the response.00:51:48.001 --> 00:51:51.009


So we need to do some kind of fine tuning.00:51:51.009 --> 00:52:00.842


But instead of going through all the documentation, videos, transcripts, I asked what are the different statuses, what are the major modules?00:52:00.842 --> 00:52:03.385


It was able to give me all that information.00:52:03.385 --> 00:52:10.565


That was really, you know, reduced my work at least 200 hours just to go through the documentation.00:52:10.625 --> 00:52:12.208


It's incredible the use.00:52:12.208 --> 00:52:33.664


I just get excited and I can go off on tangents because think, now, having all that information readily available, you also don't need to memorize and if you can retrieve the information quickly, now you can prompt to get the information back without having to spend time searching or memorizing or going through.00:52:33.664 --> 00:52:50.509


Nothing's perfect, but even if it can get you 80% of the way there, or even something as simple as you had mentioned, that it can do surprisingly well most of the steps or most of the things that are outlined, at least it gets you started and it can show you the reference documents where you can look more.00:52:50.509 --> 00:53:03.860


I'm waiting for the day where I just it's like we plug in the microphone jack into a computer to speak that you have a little jack you plug into your head and you just think and all of this information will come into your brain and you know.00:53:03.860 --> 00:53:07.302


It's like having an external hard drive Exactly, you know.00:53:07.322 --> 00:53:12.706


What's interesting is that you gave this a perfect example on a specific industry that's.00:53:12.706 --> 00:53:17.349


You know, even in legal it's the same thing If you ever dealt with.00:53:17.349 --> 00:53:24.938


There's a discovery and there's like tons and tons of documents and typically a paralegal would be the one that's doing that.00:53:24.938 --> 00:53:30.059


Right, they would have to reference a specific document, and it's a lot of writing and all this stuff.00:53:30.059 --> 00:53:38.985


Now you can just use Copilot to do that and be able to summarize or even search for a specific thing like, hey, did this person say this?00:53:38.985 --> 00:53:45.880


And then they'll say yes, they did for that particular topic, and then it's going to reference where it found that on the documentation.00:53:45.880 --> 00:53:49.655


So that's actually a perfect use case.00:53:51.079 --> 00:54:07.947


Sorry, and it goes back to the co-pilot that we were talking about, where a co-pilot is someone that can when I say someone, see, I'm referring it, now, it's like a human being someone that can, when I say someone, see, I'm referring it, now, it's like a human being, it's a tool that coordinates all the other agents.00:54:07.947 --> 00:54:19.650


Another perfect use case for that would be if a client interacts with your co-pilot and says, hey, I want to look at your products that are available based on my description.00:54:19.650 --> 00:54:20.054


I'm asking.00:54:20.054 --> 00:54:37.141


Then it could look at your data database and looks at hey, these are the items available for this business and then if the client's interested in purchasing that, then that same co-pilot can call another agent to create a sales order.00:54:37.141 --> 00:54:41.025


So you got to look at that way.00:54:41.025 --> 00:54:44.221


When we're talking about co-pilots, that's another use case.00:54:44.221 --> 00:54:55.969


When it's calling multiple agents based on different tasks One's informational and then the other one could be a task where it creates a sales order and then submits it to maybe a business central.00:54:55.969 --> 00:55:00.172


So many different ways, use cases, so many different ways.00:55:00.295 --> 00:55:01.077


Use cases.00:55:01.077 --> 00:55:07.309


Yeah, that legal area is legal vertical, especially legal industry, like healthcare is also.00:55:07.309 --> 00:55:12.443


So I was very interesting this AI is going to disrupt more.00:55:12.443 --> 00:55:19.128


I was reading somewhere about which area AI is going to have a very quick ROI.00:55:19.128 --> 00:55:27.701


Looks like it's interestingly a financial sector, because everything is really predictable, mathematical.00:55:27.701 --> 00:55:39.463


So what they are saying is in the financial sector AI will have a lot of impact, means it can bring a lot of ROI In terms of healthcare.00:55:39.463 --> 00:55:49.534


It can unlock a lot of new medicines and solve and come up with a lot of breakthroughs in the healthcare sector.00:55:49.534 --> 00:55:53.639


So, as in legal, which is more on the documentation side.00:55:53.639 --> 00:56:01.929


You are right, Chris, the paralegal work, like meeting notes, paralegal work also could be now backside.00:56:01.929 --> 00:56:08.706


You know, take a back step, or paralegals can use the copilot to come up with their understanding and revalidate.00:56:08.954 --> 00:56:20.206


Finance is interesting, though, because it's a large, natural, large language model, but it doesn't does it do a good job crunching numbers, though, if it gives you a bunch of data.00:56:21.657 --> 00:56:23.639


I think you need to have the right tool for the job.00:56:23.639 --> 00:56:27.465


That this is where it goes back to.00:56:27.465 --> 00:56:32.260


If you're trying to use a hammer to do math, it's not going to work.00:56:32.260 --> 00:56:35.067


If you're going to use a calculator to do math, it's going to work.00:56:35.067 --> 00:56:40.391


And this is where I think humans have this natural ability to want to destroy.00:56:40.391 --> 00:56:41.235


You build a robot.00:56:41.235 --> 00:56:42.717


People will start throwing things at it.00:56:42.717 --> 00:56:42.998


Why?00:56:42.998 --> 00:56:46.226


Because Because it's a robot, and I think the same thing.00:56:46.266 --> 00:56:51.686


When this first came out, everyone's like oh, I trained it to do four plus four equals nine and not eight.00:56:51.686 --> 00:56:54.219


It's silly, but you have to step back.00:56:54.219 --> 00:56:55.923


That's not what it's supposed to be doing.00:56:55.923 --> 00:57:01.201


It's supposed to retrieve and summarize and show information, not do math.00:57:01.201 --> 00:57:02.322


So even go to the agent.00:57:02.322 --> 00:57:08.460


The agent that sends the email isn't going to be the same agent that responds to the email in our scenario.00:57:08.460 --> 00:57:11.583


So it's a matter of using the right tool.00:57:12.054 --> 00:57:15.184


You mentioned model Sai or the right agent for the job.00:57:15.184 --> 00:57:31.306


So I think, with finance, depending upon listen, finance, if you're talking investments and finance that whole market can spiral if you just let it go, ai out of control, because it will just read the patterns and respond to the patterns and you could just have a sharp crash or a sharp spike.00:57:31.306 --> 00:57:44.751


But I think, from an organizational point of view, you can use AI to help with financial information, financial reports or even some analysis of information, which you need the right agent and you also need to have the proper data.00:57:44.751 --> 00:57:52.827


This topic can get me all of it when it comes to AI within the world.00:57:52.827 --> 00:58:15.742


It's great, but to go back to now, business Central, to go back to Copilot, agent Power Platform and the use of AI within that space is where I really like to focus, because I think, brad, the enterprise, especially enterprise, ai, Microsoft, the way I assessed a couple of years back, I think that still holds good.00:58:16.583 --> 00:58:36.224


The feature also looks like Microsoft is going to win a lot of Dynamics 365, er, pcrm and go head-to-head with Salesforce, hubspot, oracle all the spaces given their ecosystem, microsoft ecosystem, especially with Cloud, azure, with data.00:58:36.224 --> 00:58:38.483


They streamlined their data platform.00:58:38.483 --> 00:58:40.722


Now they are calling it as a Microsoft fabric.00:58:40.722 --> 00:58:41.864


So that is a.00:58:41.864 --> 00:58:49.407


Satya told that that is the greatest enhancement that they did to their SQL Server data platform.00:58:49.407 --> 00:58:51.481


Right, sql Server, power BI.00:58:51.481 --> 00:59:00.005


They have ETLs I work with their ETLs like Azure Data Factory and SSIS, ssrs, all those things you know.00:59:00.005 --> 00:59:03.916


They all club together and now they are calling as a Microsoft Fabric.00:59:03.916 --> 00:59:08.751


Data is one of the very important things for these co-pilots or agents to work.00:59:08.751 --> 00:59:12.461


So Microsoft really, you know, nailed it.00:59:12.461 --> 00:59:13.605


It is competing.00:59:13.605 --> 00:59:20.784


I have some experience with Snowflake also, which can go to different clouds and gather the information.00:59:20.784 --> 00:59:21.585


That's what.00:59:21.967 --> 00:59:25.114


I was a data architect for a couple of years.00:59:25.114 --> 00:59:32.969


In 2017 to 2019, I worked for Ford and FedEx as a solution architect.00:59:32.969 --> 00:59:36.164


I started with the programming language.00:59:36.164 --> 00:59:42.420


Then, in one of the meetings, one of the project managers said, hey, that is a matrix organization.00:59:42.420 --> 00:59:42.585


I asked what is it?00:59:42.585 --> 00:59:43.193


Project managers said hey, that is a matrix organization.00:59:43.193 --> 00:59:44.239


I asked what is it?00:59:44.239 --> 00:59:46.041


He said, hey, that is a PMP thing.00:59:46.041 --> 00:59:47.981


So I said I'm interested to learn.00:59:47.981 --> 00:59:48.563


Can you help me?00:59:48.563 --> 00:59:49.639


He said go and get PMP.00:59:49.639 --> 00:59:55.802


So I went and I got PMP and in almost 2009,.00:59:55.802 --> 00:59:57.045


They gave me a big project.00:59:57.045 --> 00:59:58.027


I ran the project.00:59:58.027 --> 01:00:01.597


I understand how to run the projects and what it is.01:00:01.597 --> 01:00:04.000


I got real good respect for the project managers.01:00:04.000 --> 01:00:09.423


Then I looked at the space and architecture is the technology thing that I like.01:00:09.423 --> 01:00:11.626


So I worked as an architect.01:00:11.626 --> 01:00:19.251


I went and I got my TOGAF Enterprise Architecture Certification to look at the space, this IT, differently.01:00:19.251 --> 01:00:25.972


So that gave me a different lens when I look at all these things and put this AI journey.01:00:25.992 --> 01:00:32.518


Microsoft is having Microsoft Fabric ecosystem, which is a data which is core to this co-pilot.01:00:32.518 --> 01:00:37.360


It's a knowledge base, and they streamlined their security landscape.01:00:37.360 --> 01:00:42.980


They have Azure or Azure KD and all those things.01:00:42.980 --> 01:00:45.282


Now they are calling it as Azure Entra.01:00:45.282 --> 01:00:46.820


So Copilot is there.01:00:46.820 --> 01:00:57.862


Now if people have Copilot enabled in their Azure security, they can go on just prompt it and say that, hey, what are the security risks?01:00:57.862 --> 01:01:02.536


There are a lot of tools, from Microsoft to Microsoft Defender and all those things.01:01:02.536 --> 01:01:04.496


So Microsoft platform, you know, especially the implementers or partners.01:01:04.496 --> 01:01:06.398


There are a lot of tools from Microsoft to Microsoft Defender and all those things.01:01:06.398 --> 01:01:09.920


So Microsoft platform, you know, especially the implementers or partners are going to win big in this year.01:01:09.920 --> 01:01:12.961


That's how I am seeing, given the way they are able to.01:01:13.121 --> 01:01:26.628


The co-pilot is integrated into all different areas of Microsoft M365, you know, licensing, security, data, dynamics 365, erp, crm, bc and they created the platform.01:01:26.628 --> 01:01:33.271


See, one of the things that Satya did was that we escorted that IDE and GitHub.01:01:33.271 --> 01:01:35.552


All these things are so tied together.01:01:35.552 --> 01:01:41.760


I think this year is going to be a very good year for Microsoft.01:01:41.760 --> 01:01:52.780


In fact, in one of the meetings I think Satya was saying that he was surprised to see Dynamics 365 is winning a lot of bids, you know, in their sales.01:01:52.954 --> 01:02:02.041


I think that's where the power of this co-pilot agents are going to be for this enterprise and implementers, especially business central space.01:02:02.041 --> 01:02:12.360


If you look at right in the SMB space, according to Gartner and Magic Quadrant, there are very few competing companies right, netsuite, bc.01:02:12.360 --> 01:02:22.467


So in that also, oracle NetSuite is not having the kind of ecosystem that Microsoft is having right, so they don't have the Copilot, they don't have the Azure Cloud.01:02:22.467 --> 01:02:23.373


They don't have the Azure cloud.01:02:23.373 --> 01:02:25.983


They have OCI, oracle Cloud Interface.01:02:25.983 --> 01:02:31.909


They are still evolving, whereas Microsoft they have copilot, their UI and the backend it's the agents.01:02:31.909 --> 01:02:32.914


I think it is one of the really good years.01:02:32.934 --> 01:02:33.788


They can use the connector.01:02:33.788 --> 01:02:50.074


Well, there's a lot to copilot and you're mentioning the ecosystem from the Microsoft platform, which to me you mentioned that fabric is sort of a blend of a lot of different separate services now into one level plane.01:02:50.074 --> 01:02:58.210


I think the whole ecosystem is changing as well, where you have the ERP interface, the ERP software, the data backend, the automation, the tasks.01:02:58.210 --> 01:02:59.737


There's a lot to this.01:02:59.737 --> 01:03:16.836


Copilot within the Microsoft ecosystem we talked about co-pilot studio, we're talking about agents, we're talking prompting when is the best place for someone to go to learn?01:03:16.836 --> 01:03:19.461


And also, how much do you learn?01:03:19.461 --> 01:03:23.409


Right, I mean it's I drive a vehicle.01:03:23.409 --> 01:03:26.465


Do I need to know how to build the vehicle?01:03:26.465 --> 01:03:27.940


Do I need to know how to fix the vehicle?01:03:27.940 --> 01:03:31.302


I just didn't know to go to talk to somebody, but I still need to learn how to drive the vehicle.01:03:31.302 --> 01:03:38.041


Or you know where do you go to do what, but Brad, that's like this journey.01:03:39.556 --> 01:03:42.695


That's like back in the day when someone says I'm in IT, right.01:03:42.695 --> 01:03:44.961


That's like back in the day when someone says I'm in IT, right.01:03:44.961 --> 01:03:47.940


Back then it was like, oh, you fix computers, but now that's not the case, right?01:03:47.940 --> 01:03:52.440


When you say I'm in IT, it's like which part If anyone tells me they're in IT?01:03:52.621 --> 01:03:56.282


Chris, I don't even talk to them anymore, because if someone says well, what do you do?01:03:56.282 --> 01:04:11.382


And they say I'm in IT To me right there, that just means I'm not going to talk to you, because you think you do everything and you really don't, because it's impossible to know everything in IT.01:04:11.402 --> 01:04:17.871


It's almost like AI, because AI itself encompasses much, much more than what people think it does with just saying a large language model.01:04:17.871 --> 01:04:33.994


But where are some?01:04:33.994 --> 01:04:37.739


Where can someone go to start to learn their journey of maybe creating an agent to do emails for a podcast, or maybe even help schedule their calendar time, or I don't even know?01:04:37.739 --> 01:04:46.590


I'm trying to think of all the other practical uses you could use, both professionally and personally, to have some of these agents simplify your day so that you have more time to do the things that are enriching you in your life and your family's lives.01:04:47.931 --> 01:04:52.726


Yeah, sure, I think Microsoft documentation is the first place to go.01:04:52.726 --> 01:04:59.159


I always refer that learnmicrosoftcom there for this co-pilot.01:04:59.159 --> 01:05:00.784


There is a very good documentation.01:05:00.784 --> 01:05:03.643


Microsoft really improved over the few years.01:05:03.643 --> 01:05:10.849


That is the best place and to know complete details, technical details about it.01:05:10.849 --> 01:05:17.463


And if you have a question in the Microsoft copilot studio and all those things, go to the forums.01:05:17.463 --> 01:05:24.264


Microsoft is having good forums, dynamics, our Power Platform, community forum from Microsoft.01:05:24.264 --> 01:05:31.567


I answer a lot of questions in that, so that's another good place to get quick answers.01:05:31.655 --> 01:05:32.478


Oh, I'm going direct to you.01:05:32.539 --> 01:05:34.782


after this, we're going to have a lot of conversations.01:05:34.782 --> 01:05:36.922


You do a newsletter too, right?01:05:36.922 --> 01:05:38.233


That's what I was just going to mention.01:05:40.581 --> 01:05:45.094


I publish weekly a newsletter called D365 Co-Pilot Digest on LinkedIn.01:05:45.094 --> 01:05:51.668


In that, my goal is to just give a quick update on four things.01:05:51.668 --> 01:05:54.905


Number one, dynamics 365, erp or CRM.01:05:54.905 --> 01:05:56.920


That is first one.01:05:56.920 --> 01:06:01.061


What are the updates that Microsoft is publishing Are they releasing any new co-pilots?01:06:01.061 --> 01:06:03.181


And second one is Power Platform how the Power Platform is evolving, given Microsoft is publishing are they releasing any new copilots?01:06:03.181 --> 01:06:13.362


And second one is Power Platform how the Power Platform is evolving, given Microsoft is integrating AI into copilot, into their Power Platform source, like Power Apps and Power Automate.01:06:13.362 --> 01:06:17.626


So that is the second topic I write weekly on the Digest.01:06:17.626 --> 01:06:21.679


And third one is Copilot Studio or AI, microsoft AI related things.01:06:21.679 --> 01:06:23.510


And fourth one is Microsoft Fabric, or, uh, yeah, I, microsoft, yeah, I related things.01:06:23.510 --> 01:06:26.862


And fourth one is microsoft fabric, microsoft data platform related.01:06:26.862 --> 01:06:30.655


I feel like these are the core for this next evolution.01:06:30.655 --> 01:06:33.760


So I write this uh, linkedin digest.01:06:33.760 --> 01:06:36.407


You can find that digest on the linkedin I.01:06:36.407 --> 01:06:39.159


You know I can share that information we'll also put the.01:06:39.581 --> 01:06:53.824


we have a guest page now, so on the episode on the website, we'll also put the profile, which has a link to your LinkedIn profile, so that someone could read past issues of your digest and also see the new issues that are coming out.01:06:53.824 --> 01:06:59.086


So, cy, my mind is blown again this AI thing.01:06:59.086 --> 01:07:05.358


I don't know where to begin and where to end with it, but You're still confused.01:07:05.358 --> 01:07:10.786


It's not that I'm confused, I understand it, you know.01:07:10.786 --> 01:07:27.340


And some things I say in jest, because it's just so much, so fast, that I think it's important to find the nugget that interests you or you think that you'll benefit from and be aware of the others and pursue that.01:07:27.902 --> 01:07:30.447


It's almost like the master of what is it?01:07:30.447 --> 01:07:32.039


Jack of all trades, master of none.01:07:32.039 --> 01:07:48.715


I think you need to start focusing on where you think you'll get the best ROI for what you're going to do and then be aware of the other stuff, because you may have to change your thought process because of something else that you can incorporate to what you want to do.01:07:48.715 --> 01:07:59.755


So I'm not confused, I'm overwhelmed and exciting, and I could go down so many different tangents, because AI itself has so many different roads that you can go down side roads or side streets.01:07:59.795 --> 01:08:03.905


I guess you could say yeah, broadly two, two, two areas.01:08:03.905 --> 01:08:08.485


Right one is b2c space and under is enterprise b2b space.01:08:08.485 --> 01:08:13.838


So in the b2c, b2c space there is so much happening, you're right who are going to be players?01:08:13.838 --> 01:08:17.686


Especially deep seek model is able to.01:08:17.686 --> 01:08:19.719


You know the the thing that happened.01:08:20.140 --> 01:08:26.582


It's also very interesting the new I need to get one of those, and we'll have an episode coming up shortly to talk about that.01:08:26.582 --> 01:08:31.603


But I need to get an lm installed locally, or do I even need to right that's my question.01:08:31.603 --> 01:08:38.426


It's do I need to have it or can I just use one of the existing tools and models and go from there?01:08:38.487 --> 01:08:48.284


I think that's a whole other question it is like you know your personal laptop versus having a vm on the cloud right, so do you.01:08:48.284 --> 01:08:49.127


Which one you prefer?01:08:49.576 --> 01:09:06.484


well, it's a matter of mac os well, si, thank you, I could talk with you for days and, uh, oh, you're going to hear from me shortly after this because, chris, we're going to get some emailing set up, hopefully even some.01:09:06.484 --> 01:09:07.748


Yeah, we should build one out.01:09:07.748 --> 01:09:09.573


Yeah, even even some.01:09:09.654 --> 01:09:13.403


Oh no, sorry, you volunteered did you hear that we have it on recording?01:09:13.403 --> 01:09:14.586


Definitely.01:09:14.666 --> 01:09:25.875


So yeah, we'll get some emailing ai set up that's in in some fashion to assist with the scheduling, to give us the opportunity to speak with more guests and not have to schedule.01:09:25.895 --> 01:09:30.766


Yeah, if you guys are doing manually, we have to optimize that.01:09:30.766 --> 01:09:34.405


You know, the four-hour work week book is the best one.01:09:34.405 --> 01:09:42.122


You know, when you try to do this kind of really at scale, try to automate it and optimize it so that you can just move on.01:09:42.194 --> 01:09:47.384


That's where we're looking to go, but again, thank you again for all the information you shared.01:09:47.384 --> 01:09:50.658


Thank you for all that you do for the microsoft ecosystem.01:09:50.658 --> 01:09:52.201


You share a lot of information.01:09:52.201 --> 01:09:59.252


I read your digest as well, as I do come across you at some points, uh, within the forum, seeing some of this as I'm reading up on it.01:09:59.252 --> 01:10:00.636


So we appreciate that, all that you do.01:10:00.636 --> 01:10:10.364


In the meantime, if anybody would like to get in contact with you to learn more about AI and some of the approaches that are available, what's the best way to contact you?01:10:11.855 --> 01:10:14.623


Yeah, linkedin, sai Thurlapati.01:10:14.623 --> 01:10:17.591


Or, like you mentioned, there will be a page.01:10:17.591 --> 01:10:19.537


I will share that information with you.01:10:19.537 --> 01:10:24.664


Or D365, copilot Digest is the other newsletter.01:10:24.664 --> 01:10:25.546


Those are the two ways.01:10:25.565 --> 01:10:26.328


Excellent, excellent.01:10:26.328 --> 01:10:27.029


Thank you again.01:10:27.029 --> 01:10:28.010


We appreciate your time.01:10:28.010 --> 01:10:29.981


Look forward to speaking with you very soon.01:10:30.675 --> 01:10:33.564


Very, very soon, thank you.01:10:33.564 --> 01:10:34.286


Thank you so much.01:10:34.286 --> 01:10:36.198


I listened to your episodes.01:10:36.198 --> 01:10:37.121


I learned so much.01:10:37.180 --> 01:10:37.762


Thank you.01:10:37.762 --> 01:10:39.466


Thank you, we appreciate it.01:10:39.466 --> 01:10:40.494


Have a good day, ciao, ciao.01:10:40.614 --> 01:10:41.076


Bye for now.01:10:41.076 --> 01:10:41.997


You too Bye.01:10:43.177 --> 01:10:50.305


Thank you, chris, for your time for another episode of In the Dynamics Corner Chair, and thank you to our guests for participating.01:10:50.586 --> 01:10:52.127


Thank you, brad, for your time.01:10:52.127 --> 01:10:55.570


It is a wonderful episode of Dynamics Corner Chair.01:10:55.570 --> 01:11:00.359


I would also like to thank our guests for joining us.01:11:00.359 --> 01:11:02.105


Thank you for all of our listeners tuning in as well.01:11:02.105 --> 01:11:16.605


You can find Brad at developerlifecom, that is D-V-L-P-R-L-I-F-E dot com, and you can interact with them via Twitter D-V-L-P-R-L-I-F-E.01:11:16.605 --> 01:11:31.578


You can also find me at matalinoio, m-a-t-a-l-i-n-o dot I-O, and my Twitter handle is matalino16.01:11:31.578 --> 01:11:33.684


And you can see those links down below in the show notes.01:11:33.684 --> 01:11:35.028


Again, thank you everyone.01:11:35.028 --> 01:11:38.119


Thank you and take care.
Sai Turlapati Profile Photo

Sai Turlapati

Microsoft MVP | Dynamics 365 Copilot/Agent