Three years ago: All in Ai

1 year ago: All in LLM

Now: All in Agent

Course planning: 100+courses , continuously updated

Supporting materials: provide all courseware and case materials

• What is an agent: do an Ai to work for you

• Why don't you need to learn any Ai tools

• How to define the role (it is an expert of XXX just by prompt words?)

• How to design business processes for agents

• Is there only one agent?

1、 Learning gains

1. Everyone's AI=own needs+business scenarios+tool use+standardized processes

2. The course teaches everyone to integrate: take the big model as the brain and let it use various tools according to the process

3. You can make your own Ai product by any DIY and combining your own needs, scenarios and processes

4. Provide all case templates, first run the process, and then replace to complete their own tasks

5. The courses are continuously updated, and more cutting-edge technologies and cases will be continuously added to everyone in 24 years

6. Timeless technology, tool software will be replaced, and the agent can configure skills at will

2、 Chapter content

Chapter 1 Interpretation and Application Analysis of Agent Architecture

[Topic of this chapter] Popular explanation of the principle and architecture of agents

1-Analysis of problems to be solved by Agent 2-Basic capabilities required by Agent

3 - Analysis of the relationship with the large model 4 - Analysis of the definition of multi-agent

5 - The role of the framework and the problems that can be solved 6 - Overall summary and analysis

7-GPTS Analysis A Wave 8-Classical Task Analysis

Chapter 2 GPTS Building Agent Practice

[Topic of this chapter] Use GPTS to build agent modules and integrate corresponding actions

1-Overview and analysis of GPTS task process 2-Control method of calling API

3-API related configuration completed 4-Complete instructions and scripts and generate them

Chapter 3 Create exclusive customer service in your own field

[Topic of this chapter] Integrate the big model and various APIs from scratch to complete customer service functions

1-DEMO demonstration and overall architecture analysis 2-backend GPT project deployment launch

3 - Front end assistant API and flow chart configuration 4 - Method and process of accessing external API

5 - Introduction of API method interpretation 6 - Instruction prompt construction

7-GPTS Analysis A Wave 8-Classical Task Analysis

Chapter 4 AutogenStudio Agent Framework Practice

[Topic of this chapter] Use a localizable framework to build your own agent tasks

1-AutoGenStudio framework installation and introduction 2-Action API configuration method

3-Domestic Common API Configuration Method 4-API Interface Online Test

5 - Workflow Configuration 6 - Execution Process and Results

7-Ollama Environment Configuration and Installation 8-autogen Access Local Model

9-API generation method 10 GroupChat module

11 - Execute process analysis 12 - External local support library configuration method

13 - Add RAG Skill 14 LMStudio Local Download Deployment Model

15 Call local model method and configuration

Chapter 5 Interpretation of MetaGpt Agent Framework

[Topic of this chapter] Popular interpretation of generated code/game framework MetaGpt

1 - Overview and analysis of the paper 2 - Introduction to the overall framework logic

3 - Project environment configuration 4 - Basic interpretation - Action definition method

5 - Basic interpretation - Role definition 6 - Implementation method of single action agent

7 - Multi action configuration method 8 - Timer task environment configuration

9 - Interpretation and analysis of timer task flow

Chapter 6 MetaGpt Application Practical Application

[Topic of this chapter] Use MetaGpt to define roles and actions and complete corresponding functions

1 - Composition of basic agent 2 - Tasks and business logic definition to be completed by agent

3 - Problem Disassembly and Implementation Process 4 - Retrieve Important URLs

5 - Sub problem generation summary result 6 - Summary and result output

Chapter 7 Analysis and Application of RAG Retrieval Architecture

[Theme of this chapter] Interpretation of the role and effect of RAG and its optimization methods

1-Interpretation of tasks to be completed by RAG 2-Interpretation of RAG overall process

3 - Analysis of recall optimization strategy 4 - Interpretation of recall improvement plan

5 - Assessment tool RAGAS 6 - External local database tool

Chapter 8 Structure and Project Interpretation of Stanford AI Town

[Topic of this chapter] Interpretation of the paper and its functional architecture analysis combined with source code analysis

1 - Overall story interpretation 2 - Problems to be solved and overall framework analysis

3-The basic framework of the paper analyzes the memory information of 4-Agent

5 - Construction process of perception and reflection module 6 - Implementation details of planning module

7 - Overall Process Framework Figure 8 - Perception Module Interpretation

9 - Interpretation of thinking module 10 - Interpretation of project environment configuration method

Chapter 9 MOE Multi Expert System

[Topic of this chapter] Analysis of the role and effect of multi expert system combined with mistrialAi

1-MOE Overview Analysis 2-MOE Module Implementation Method Interpretation

3 - Effect analysis and summary

Chapter 10 Interpretation of LLM and LORA Fine Tuning Strategies

[Topic of this chapter] Analysis of fine-tuning methods and strategies for large models

1. How to do downstream tasks for large models 2. LLM landing micro adjustment analysis

3-LLAMA and LORA introduce the core idea of 4-LORA fine tuning

5-LORA model implementation details

Chapter 11 LLM Downstream Task Training Model Practice

[Topic of this chapter] Fine tune the big model based on actual tasks and data sets

1 - Function of prompting project 2 - Interpretation of project data

3 - Source code calling DEBUG interpretation 4 - Training process demonstration

5 - Effect demonstration and summary analysis

Chapter 12 OPENAI-LLM Model Optimization Summary

[Theme of this chapter] Analysis of optimization of large model and interpretation of common strategies

1-RAG and problems that can be solved or cannot be solved by fine-tuning 2-RAG practice strategy

3 - Problems to be solved by fine-tuning

Chapter 13 Supplement: Interpretation of the whole process of computer vision project

[Theme of this chapter] Whole process interpretation of how to do visual projects and how to implement each step

1 - Project demand analysis process 2 - Data and feature database preparation

3 - Model preparation and project analysis 4 - Summary of model selection methods

5 - Project experience summary and optimization method

Chapter 14 Supplement: Interpretation of the whole process of knowledge mapping project

[Theme of this chapter] Whole process interpretation of how to do knowledge mapping projects and how to implement each step

1 - Analysis of problems and processes to be solved by knowledge mapping 2 - Analysis of practical application of knowledge mapping project

3 - Interpretation of practical application projects of knowledge map 4 - Problems to be solved by the big model and application analysis

5 - Summary and analysis of tools

Chapter 15 Supplement: Interpretation of the Whole Process of Data Mining Project

[Topic of this chapter] The whole process explains how to do data mining projects and how to implement each step

1 - Problems to be solved by data mining 2 - Data processing and cleaning analysis

3-Function and process of feature engineering 4-Analysis of machine learning algorithm

5 - Where to find the template