Chapter 1: LLM Basics & Ollama Practical
- 1.1 Core Concepts & Value of Large Language Models
- 1.2 LLM Parameters & Technical Essence
- 1.3 Transformer Architecture Deep Dive
- 1.4 LLM Data Training & Compute Logic
- 1.5 Why LLMs Must Rely on GPUs
- 1.6 Ollama CLI Basic Operations
- 1.7 Ollama Model Start & Stop Full Workflow
- 1.8 Ollama Interactive Dialogue & Function Testing
- 1.9 Ollama Model Download & Version Management
- 1.10 Ollama Offline Deployment: Dependencies & Installation
- 1.11 Ollama Online Deployment: Network & Version Configuration
- 1.12 Ollama Deployment Verification & Service Status Check
- 1.13 Common Ollama Deployment Issues & Solutions
- 1.14 Ollama Config Files & Custom Runtime Rules
- 1.15 Ollama CPU/GPU Resource Limits & Control
- 1.16 Ollama Log Analysis & Troubleshooting
- 1.17 Ollama API Interface Practical Calls
- 1.18 Ollama High-Availability Deployment Plan
- 1.19 Ollama & AIOps Scenario Fit Advantages
Chapter 2: Model Selection & DeepSeek Practical
- 2.1 Open-Source LLM Ecosystem & Characteristics
- 2.2 Closed-Source LLM Ecosystem & Characteristics
- 2.3 Enterprise LLM Selection Core Dimensions
- 2.4 Cross-Industry Scenario Model Selection Strategy
- 2.5 DeepSeek Model Deployment & Verification Steps
- 2.6 DeepSeek Q&A Function Practical Testing
- 2.7 DeepSeek Service Status View & Monitoring
- 2.8 DeepSeek Resource Shortage Response Plan
- 2.9 DeepSeek Startup Parameters & Performance Tuning
- 2.10 DeepSeek API Interface Practical Integration
- 2.11 DeepSeek & Enterprise Knowledge Base Integration Practical
- 2.12 DeepSeek Docker Containerized Deployment Plan
- 2.13 Ollama Deployment DeepSeek Full Workflow
- 2.14 Unified Model Deployment in AIOps Applications
Chapter 3: Enterprise LLM Private Deployment & Hardware Architecture Design
- 3.1 Hardware Configuration Estimation
- Hardware Core Configuration Logic
- Quantization Tech & VRAM Optimization
- Typical Budget Plans
- VRAM Estimation Formula
- 3.2 Hardware Selection Strategy
- Demand-Driven Decision Making
- Typical Enterprise Scenario Cases
- 3.3 Open-Source LLM Deployment Practical
- Deployment Ideas & Plans
- Alibaba Cloud PAI One-Click Deployment
- Tencent Cloud HAI Experience
- Ollama Deployment Method
- VLLM Deployment Method
- Cluster Mode Deployment
Chapter 4: LLM Fine-Tuning & Enterprise Custom Model
- 4.1 Understanding LLM Fine-Tuning
- Why Fine-Tuning Is Needed
- Fine-Tuning Tech Classification
- Tech Option Guide
- Fine-Tuning Strategy
- 4.2 LLM Fine-Tuning Tools
- Open-Source Fine-Tuning Tools
- Commercial Fine-Tuning Platforms
- 4.3 LLM Fine-Tuning Datasets
- Fine-Tuning Dataset Classification
- Dataset Formats
- Acquire Public Datasets
- Create Your Own Dataset
- 4.4 Fine-Tuning Hyperparameters
- 4.5 LLM Fine-Tuning Practical
- iFlytek Spark Fine-Tuning Practical
- LLaMA-Factory Fine-Tuning Qwen3 LLM
- Unsloth Fine-Tuning Qwen3 LLM
Chapter 5: Intelligent Agent System Construction & Automated Ops Practical
- 5.1 Intelligent Agent Theory Foundation
- 5.2 Coze Intelligent Agent Platform
- 5.2.1 Quick Experience Coze Agent
- 5.2.2 Workflow Practical
- 5.2.3 Dialogue Flow Practical
- 5.2.4 Knowledge Base Practical
- 5.2.5 Database Practical
- 5.3 Coze Agent Platform Practical Deployment
- 5.3.1 Create Workflow
- 5.3.2 Create Agent
- 5.4 Build Agent with Open Source Dify Platform
- 5.4.1 Introduction to Dify
- 5.4.2 Local Deployment of Dify
- 5.4.3 Configure Model in Dify
- 5.4.4 Configure Plugins in Dify
- 5.4.5 Create Chatflow App
- 5.4.6 Create Workflow
- 5.4.7 Create Knowledge Base
- 5.4.8 Dify Agent Practical
- 5.5 MCP-Based (Alibaba Cloud Bailian) Agent Practical
- 5.5.1 MCP Basics
- 5.5.2 Quick Experience Alibaba Cloud Bailian MCP
- 5.5.3 Build a Travel Planner with Alibaba Cloud Bailian MCP
- 5.6 Private Deployment of Open-Source Coze
- 5.6.1 Prepare Linux Machine
- 5.6.2 Install Docker & Docker-compose
- 5.6.3 Clone Source Code
- 5.6.4 Run Coze
- 5.6.5 Use Coze
- 5.7 n8n-Based Agent Practical
- 5.7.1 N8n Introduction
- 5.7.2 N8n Deployment
- 5.7.2.1 Preparation
- 5.7.2.2 Install n8n
- 5.7.3 Experience n8n
- 5.7.3.1 Create Workflow Based on Template
- 5.7.3.2 Custom Workflow
- 5.7.3.3 n8n Node Introduction
- 5.7.3.3.1 Trigger Node
- 5.7.3.3.2 File Operation Node
- 5.7.3.3.3 Control Node
- 5.7.3.3.4 Loop & Iteration
- 5.7.3.3.5 Merge
- 5.7.3.3.6 Flow Control
- 5.7.3.3.7 Code Node
- 5.7.3.3.8 Data Node
- 5.7.3.3.9 Storage Node
- 5.7.3.3.10 Third-Party Storage
- 5.7.4 Build Agent with n8n
Chapter 6: RAG Retrieval Augmentation & Enterprise Knowledge Base Deployment
- 6.1 RAG Basics
- 6.2 Vector Database Milvus
- 6.2.1 Understanding Vector Databases
- 6.2.2 Quick Start Milvus
- 6.3 RAG Implementation with FastGPT
- 6.3.1 FastGPT Introduction & Installation
- 6.3.2 Quick Start FastGPT
- 6.3.3 Project Practical
- 6.4 RAG Implementation with RAGFlow
- 6.4.1 Introduction to RAGFlow
- 6.4.2 Deploy RAGFlow on Linux
- 6.4.3 Quick Experience RAGFlow
- 6.4.4 Project Practical
Chapter 7: LLM Ops Monitoring, Performance Tuning & Security Governance
- 7.1 LLM Platform Monitoring
- 7.1.1 Basic Command Line Tools
- 7.1.2 Professional Monitoring Tools Prometheus+Grafana
- 7.2 LLM Optimization
- 7.2.1 Optimization Strategy
- 7.2.2 LLM Quantization
- 7.2.3 LLM Knowledge Distillation
- 7.2.3.1 Core Mechanism of Knowledge Distillation
- 7.2.3.2 Knowledge Distillation Tech Method Classification
- 7.2.3.3 Baidu AI Cloud Qianfan LLM Platform for Distillation
- 7.2.3.4 Use DistillKit for LLM Distillation
- 7.3 LLM Stress Testing
- 7.3.1 Stress Test Metrics
- 7.3.2 Stress Test Tools
- 7.3.2.1 Alibaba Cloud PAI Model Online Service (EAS)
- 7.3.2.2 Baidu AI Cloud Qianfan ModelBuilder
- 7.3.2.3 EvalScope
- 7.3.2.4 Locust
- 7.3.3 Stress Test Practical
- 7.4 LLM Security Ops
Chapter 8: AIOps Intelligent Ops Platform & Enterprise AI Hub Practical
🔥 AIOps is the key chapter of this course, with the most content, ongoing updates
- 8.1 AI-Assisted Programming
- 8.1.1 GLM4.6 Full-Stack Development
- 8.1.2 Tongyi Zero-Code
- 8.1.3 Trae
- 8.1.4 Claude Code
- 8.2 Product Requirements Document Design
- 8.3 Project Development
- 8.3.1 Implement Requirements with Tongyi Lingma
- 8.3.2 Implement Requirements with Codex/Claude Code
- 8.3.3 Clone a Website with AI
- 8.4 Project Testing & Deployment
- 8.4.1 Register Account
- 8.4.2 Push Code to GitHub
- 8.4.3 Deploy Project on Vercel
- 8.4.4 Bind Domain
- 8.5 Build Ops Agent with Coze
- 8.5.1 Coze Custom Plugin
- 8.5.1.1 Create Plugin Based on API
- 8.5.1.2 Create Custom Plugin Based on IDE
- 8.5.2 Custom Coze Plugin to Manage Alibaba Cloud Machines
- 8.5.2.1 Preparation
- 8.5.2.2 Create Coze Plugin
- 8.5.3 Design Coze Workflow
- 8.5.4 Design AIOps Agent
- 8.6 Automated Ops Agent with Coze+Ansible
- 8.6.1 Preparation
- 8.6.1.1 Prepare Ansible Environment
- 8.6.1.2 Write Ansible API Service Script & Enable API
- 8.6.1.3 Write Playbook
- 8.6.2 Create Coze Plugin
- 8.6.3 Create Coze Workflow
- 8.6.4 Configure Coze Agent
- 8.7 Ops Agent with Dify+Jumpserver
- 8.7.1 Deploy Jumpserver
- 8.7.1.1 Deploy Jumpserver
- 8.7.1.2 Quick Experience Jumpserver
- 8.7.2 Deploy Jumpserver MCP
- 8.7.2.1 Get User Token
- 8.7.2.2 Deploy Jumpserver MCP
- 8.7.2.3 Add Jumpserver MCP in Dify
- 8.7.3 Implement a Simple Requirement
- 8.7.3.1 Create Dify App
- 8.7.3.2 Test Dify App
- 8.7.4 Build a Comprehensive Application Agent
- 8.8 Ops Agent with Dify+K8s
- 8.9 Ops Agent with Dify+Prometheus+Alertmanager
- 8.10 Ops Agent with n8n+Prometheus+Alertmanager
- 8.11 Ops Agent with Dify+Ansible MCP
- 8.12 DevOps+AIOps Agent with n8n+Jenkins
- 8.13 Enterprise-Level AI Agent Ops Hub Practical
- OpenClaw Introduction
- OpenClaw Deployment
- OpenClaw Chat Tool Integration
- OpenClaw Multi-Agent Collaboration
- OpenClaw Practical
- OpenClaw & AIOps Full-Scene Application
- Hermes Agent Concept Analysis
- Hermes Practical
- Hermes Chat Tool Integration
- Hermes Agent Collaboration
- Hermes Fault Troubleshooting Practical
- Hermes Agent & AIOps Full-Scene Application
Continuously updating…
