1
00:00:05,939 --> 00:00:15,060
So in this lesson, we are just going to see a graphical representation of the architecture of the professional

2
00:00:15,060 --> 00:00:17,220
Elm application we just saw.

3
00:00:21,030 --> 00:00:29,640
So this is the, uh, graphical representation of the architecture of the application.

4
00:00:29,640 --> 00:00:33,360
We just saw the insights.ai.

5
00:00:33,600 --> 00:00:40,320
So this is an application that has a back end framework based on fast API.

6
00:00:40,350 --> 00:00:47,550
This is a framework based on Python language and the seq API.

7
00:00:47,580 --> 00:00:54,120
The seq insights uh application is based on lambda index as an orchestration framework.

8
00:00:55,050 --> 00:00:59,820
The vector database is uh, it uses is Postgres.

9
00:01:00,030 --> 00:01:02,910
The back end server is render.

10
00:01:02,910 --> 00:01:11,670
Com the front end framework is Next.js, which as you know is based on react.

11
00:01:11,670 --> 00:01:15,720
And as you know, react is based on the JavaScript language.

12
00:01:15,720 --> 00:01:20,010
And the front end server is very cellcom.

13
00:01:20,990 --> 00:01:26,390
So these are the two main parts of our full stack application.

14
00:01:26,630 --> 00:01:37,490
But in the back end we have a the application connected with ChatGPT as LLM foundation model.

15
00:01:37,490 --> 00:01:45,350
For our application we have 3 or 4 external APIs uh connected as well.

16
00:01:45,350 --> 00:01:53,150
And then we are storing in the PDFs and the private data we have in the PDFs.

17
00:01:53,150 --> 00:01:54,470
In this application.

18
00:01:54,470 --> 00:02:01,280
We are storing these PDFs in S3 from Amazon Web Services.

19
00:02:01,280 --> 00:02:04,280
So this is the cloud storage okay.

20
00:02:04,280 --> 00:02:14,870
So you will see that during the next lessons we are going to learn about all these different components.

21
00:02:14,870 --> 00:02:24,110
Don't worry it looks like super complex, but it is going to be easier and more simple than you think.

22
00:02:24,110 --> 00:02:26,630
You just need to have perseverance.

23
00:02:26,630 --> 00:02:30,290
And remember the key is study and practice.

24
00:02:30,290 --> 00:02:30,950
Okay?

25
00:02:32,590 --> 00:02:37,780
So remember that you can play with the the application.

26
00:02:37,780 --> 00:02:41,890
This LM application is open source by llama index.

27
00:02:41,890 --> 00:02:45,310
So you can check it out in this URL.

28
00:02:45,310 --> 00:02:50,740
And remember that this application loads a sexy documents.

29
00:02:50,740 --> 00:02:55,630
These are financial documents and a it answers.

30
00:02:57,830 --> 00:03:00,140
Questions about these documents.

31
00:03:00,170 --> 00:03:05,300
It also provides source links and it is open source by Lamar Index.

32
00:03:05,300 --> 00:03:09,200
As you know, Lamar Index is one of the orchestration frameworks we are going to use.

33
00:03:09,200 --> 00:03:17,600
So the team at Lamar Index has created an open source, this application just to show you what you can

34
00:03:17,600 --> 00:03:19,490
do using Lamar Index.

35
00:03:20,060 --> 00:03:28,550
Remember that Lamar Index is the top alternative to Lang Chain, the other most popular orchestration

36
00:03:28,550 --> 00:03:31,430
frameworks for NLM applications.

37
00:03:33,180 --> 00:03:34,680
In the next lesson.

38
00:03:35,360 --> 00:03:39,950
We are going to start talking about some of the.

39
00:03:40,690 --> 00:03:45,550
Details of the advanced architecture of an LM application.

40
00:03:45,550 --> 00:03:53,710
So we are not going to see these details in depth is going to be just a quick initial view of them.