AI FOR OPERATIONS · 2026 edition

AIOps Vibe Coding Part 8: LLMs, Agents and End-to-End AIOps

Public preview0 videos32 curriculum items

COURSE INTRODUCTION

Course Introduction

Bring AI into operations: LLMs, RAG, fine-tuning and agents
Let AI work from knowledge and real device evidence; master model serving, fine-tuning, tool integration and controlled execution.
Sessions: L25–L30, 6 sessions of about 2.5 hours each; 10 case-based assignments. Cohort 19 schedule: 2026-12-28 to 2027-01-08.
Representative cases: OpenAI structured output and tool calling; private RAG; LLaMA-Factory fine-tuning; MCP/Skill/A2A; OpenClaw/NetClaw; a six-workflow showcase.

CURRICULUM

Course Curriculum

06

01
1. 1.1
  API and context
2. 1.2
  Structured output
3. 1.3
  Device function calling
4. 1.4
  Hosted retrieval
5. 1.5
  Agents SDK and traces
02
1. 2.1
  Models and serving
2. 2.2
  RAG concepts and ingestion
3. 2.3
  Retrieval and citations in practice
4. 2.4
  Fixed questions and fact freshness
03
1. 3.1
  Use cases and data
2. 3.2
  Configuring and starting training
3. 3.3
  Logs, artefacts and serving
4. 3.4
  Unseen questions and tool regression
04
1. 4.1
  MCP model and wrapping
2. 4.2
  Authentication and IDE calls
3. 4.3
  Organising skills
4. 4.4
  A2A routing and verification
05
1. 5.1
  Comparing frameworks one by one
2. 5.2
  LangChain logs and traces
3. 5.3
  Main platform and skill evolution
4. 5.4
  Execution control and injection checks
06
1. 6.1
  Entry points and architecture
2. 6.2
  Inspection and metrics
3. 6.3
  Logs and troubleshooting
4. 6.4
  RCA and controlled change
5. 6.5
  Wrap-up and Q&A

COURSE RECORDINGS

Course Video

Watch the public preview here or open it on the original video platform.

AIOps Vibe Coding Part 8: LLMs, Agents and End-to-End AIOps

Course Introduction

Course Curriculum

L25 Helping AI understand the network and call tools: OpenAI Responses API, Function Calling and Agents SDK

API and context

Structured output

Device function calling

Hosted retrieval

Agents SDK and traces

L26 Building your own knowledge assistant: Ollama, vLLM and Qdrant RAG

Models and serving

RAG concepts and ingestion

Retrieval and citations in practice

Fixed questions and fact freshness

L27 Training a model that knows the business: LLaMA-Factory, LoRA/QLoRA and model evaluation

Use cases and data

Configuring and starting training

Logs, artefacts and serving

Unseen questions and tool regression

L28 Connecting ops capabilities to AI: MCP, Skill and A2A

MCP model and wrapping

Authentication and IDE calls

Organising skills

A2A routing and verification

L29 Agents that troubleshoot by process: LangChain/LangGraph and OpenClaw/NetClaw

Comparing frameworks one by one

LangChain logs and traces

Main platform and skill evolution

Execution control and injection checks

L30 Seeing complete AIOps: inspection, metrics, logs, troubleshooting, alert RCA and controlled change

Entry points and architecture

Inspection and metrics

Logs and troubleshooting

RCA and controlled change

Wrap-up and Q&A

Course Video

Course recordings are being prepared.