Courses where every module ends in a real lab.

Claude Code, Level by Level

Beginner · 20 to 30 hours · First lecture free

For developers adopting Claude Code, power users automating their workflow, and CCA-F candidates who want hands-on depth behind the flashcards.

• Drive the full Claude Code workflow: explore a codebase, plan a change in plan mode, implement it, and verify it before committing
• Author CLAUDE.md hierarchies and path-scoped rules that steer Claude across a real multi-package repository
• Configure least-privilege permissions, custom slash commands, skills, and lifecycle hooks in settings.json and .claude/

Capstone: An End-to-End Automation

Cert milestones: CCA-F

1. 1level 1: Usage: Chat, Cowork, and Code1st lecture free2 projects
2. 2level 2: Prompting for Agentic Work2 projects
3. 3level 3: Context and Memory: CLAUDE.md3 projects
4. 4level 4: Workflows: Plan, Build, Verify4 projects
5. 5level 5: Customization: Skills, Hooks, Permissions4 projects
6. 6level 6: Tools and MCP4 projects
7. 7level 7: Subagents and Orchestration4 projects
8. 8level 8: Autonomy and the Agent SDK4 projects

8 levels · 64 lessons · 25 challenges · 27 projects · 20 to 30 hours

AI Engineer Course

Intermediate · 40 to 60 hours

For software engineers comfortable with Python and basic ML who want to move into building, fine-tuning and deploying language models.

• Implement a decoder-only transformer end-to-end and train it on real data
• Fine-tune open-weight LLMs with LoRA and QLoRA on a single GPU
• Build a production RAG pipeline with hybrid search, reranking, and grounded generation

Capstone: Production RAG Agent on Kubernetes

Cert milestones: NCA-GENL, NCP-GENL, NCP-AAI

1. 1module 1: How LLMs actually work3 labs
2. 2module 2: Using LLMs via API2 labs
3. 3module 3: LLM Inference & Optimization3 labs
4. 4module 4: Retrieval-Augmented Generation4 labs
5. 5module 5: Fine-Tuning & Alignment4 labs
6. 6module 6: Production Serving3 labs
7. 7module 7: Evaluation & MLOps4 labs
8. 8module 8: Agents & Tool Use7 labs

8 modules · 31 labs · 115 lessons · 49 challenges · 6 projects · 40 to 60 hours

AI Red Teaming Course: LLM & Agent Pentesting

Advanced · 30 to 45 hours

For people who build or break LLM apps: ship RAG pipelines and agents, then learn to attack your own system and harden it first.

• Map the attack surface of an LLM app, RAG pipeline, and agent, and build a harness that measures attack-success-rate
• Exploit indirect prompt injection delivered through retrieved documents, tool output, and inter-agent messages
• Poison a RAG knowledge base and exploit retrieval, embedding, and cross-tenant isolation failures

Capstone: Full LLM/Agent VAPT Engagement

1. 1module 1: AI Attack Surface & Pentest Methodology2 labs
2. 2module 2: Indirect Prompt Injection2 labs
3. 3module 3: RAG Pipeline Exploitation3 labs
4. 4module 4: Improper Output Handling3 labs
5. 5module 5: Excessive Agency & Agentic Exploitation3 labs
6. 6module 6: Agentic Supply Chain & Multi-Agent Exploitation4 labs
7. 7module 7: Model & System-Prompt Leakage3 labs
8. 8module 8: Automation, Fuzzing & Red-Team Tooling2 labs

8 modules · 22 labs · 31 lessons · 15 projects · 30 to 45 hours