AI Engineering Lab（24 周自学课程）
Start here · Browse all 24 weeks · Reference ·…
课时
- Start here
- Curriculum
- Reference
- Glossary
- Roadmap
- Start here
- The 24 week journey
- .github
- curriculum
- _templates
- AI Engineering Lab: Projects
- week-01
- Week 02: Data Engineering & SQL for AI
- Week 03: Machine Learning Fundamentals
- Week 04: Deep Learning with PyTorch
- Week 05, How LLMs Work: Tokens to Transformers
- Week 06: Prompt Engineering & the Context Window
- Week 07: RAG, Vector Search & Knowledge Graphs
- Week 08, Open Models & Local Inference: GPUs, Ollama, llama.cpp
- Week 09: Quantization & Efficient Inference
- Week 10, Fine-Tuning: LoRA, SFT & DPO
- Week 11: Evals & Error Analysis for AI Systems
- Week 12, Coding-Agent Harnesses: Claude Code, Cursor, OpenCode, DSH
- Week 13: Agentic Coding Loops & Spec-Driven Development
- Week 14, Agent Fundamentals: The Loop, Tools & Memory
- Week 15, Agent Frameworks: LangGraph & the State-Graph Model
- Week 16: Multi-Agent Systems & MCP
- Week 17: OpenClaw, Hermes & Agent Operations
- Week 18: Azure AI Foundry
- Week 19: Google Vertex AI & Gemini
- Week 20: AWS Bedrock & SageMaker AI
- Week 21, Databricks Day Zero: Unity Catalog & the Lakehouse
- Week 22, Databricks Data Engineering: PySpark, Streaming & Lakeflow
- Week 23, Databricks ML & GenAI: Training, Serving, Genie
- Week 24, Databricks Production: DABs, Governance & the Capstone
- Generated data
- reference
- agents
- AI Engineering Lab: Knowledge Base
- 01: The AI Engineering Discipline
- 02: ML & Deep Learning Fundamentals
- 03: LLM Core Concepts
- 04: Prompt & Context Engineering
- 05: RAG & Graph Engineering
- 06, Model Engineering: Quantization, Fine-Tuning & Serving
- 07: Evals & Error Analysis
- 08, Local Inference & GPUs: Running Models on Your Own Hardware
- 09: Coding-Agent Harnesses & LLM Routers
- Agents & Multi-Agent Systems
- MCP & Agent Connectivity
- 12 · Enterprise Cloud AI Platforms: Azure, Google, and AWS
- 13 · The Databricks Data Intelligence Platform at a Glance
- 14: References & Bibliography
- research
- Zorost Intelligence: Brand Notes (collected from zorost.com, Aug 2026)
- Source Repo Analysis: panteamkhh/llm-engineering-lab
- 02: Andrew Ng's AI Engineering Framework, Key References & Zorost Intelligence
- 03: Coding-Agent Harnesses, LLM Routers & Local Model Stacks
- AI Agents & Multi-Agent Systems: Research Notes
- Enterprise Cloud AI Platforms: Azure, Google, and AWS
- Databricks for AI Engineering: Zero to Hero Deep Dive
- Zorost Signal: "The AI Engineering Skills Map, turned into a training plan"
- Zorost AI Fieldwork: Notes (collected from zorost.com/ai-lab/fieldwork, Aug 2026)
- Platforms
- Azure AI Foundry: Module Guide
- AWS Bedrock: Module Guide
- Databricks: Zero to Hero
- 00 · Day Zero: Account, Workspace & CLI
- 01 · Unity Catalog: Governance for Data & AI
- 02 · Compute: Clusters, Warehouses & Serverless
- 03 · Delta Lake: The Storage Layer
- 04 · Databricks SQL (DBSQL)
- 05 · PySpark: DataFrame Data Engineering
- 06 · Lakeflow: Pipelines, Jobs & Connect
- 07 · MLflow: Experiments, Tracking & Models in UC
- 08 · Feature Engineering in Unity Catalog
- 09 · Model Training: Classic ML to Distributed
- 10: Model Serving
- 11: AI Search (Vector Search) & RAG
- 12: AI Functions & Genie
- 13: Agentic AI on Databricks
- 14: AI/BI Dashboards & Databricks Apps
- 15: Declarative Automation Bundles (DABs) & CI/CD
- 16: Governance & Security
- 17: FinOps & Cost
- 18: Certification Path & the Hero's Recap
- Capstone: ZoroLogistics Lakehouse Intelligence
- Google Vertex AI: Module Guide
- resources
- AI Engineering Lab: Skills Library
- Skill: Claude Code
- Skill: Cursor
- Skill: DeepSeek Harness (DSH)
- Skill: Ollama & llama.cpp: Running Models Locally
- Skill: OpenCode
- Skill: OpenRouter
- Zorost Agent Skills: Catalog
- agent-loop-safety
- agent-ops-handoff
- ai-output-review
- cloud-deploy-gate
- context-budget-audit
- error-analysis-50
- eval-first-development
- fine-tune-readiness
- local-model-fit
- mcp-server-craft
- prompt-suite-versioning
- rag-pipeline-checkup
- spec-first-ai-feature
- using-zorost-skills