Certify

1. 01
   D1 · Agents and Workflows — 14.7 % 8 min read · Level 2

   Agent architecture (when an agent, when a workflow, hierarchies), building with the Agent SDK or a hand-written loop, hooks, and the patterns: tool-use loops, subagents, memory.

2. 02
   D2 · Applications and Integration — 33.1 % 14 min read · Level 2

   The heaviest domain: from requirements to architecture, the mechanics of the Messages API, engineering basics (REST, JSON, async, git) and configuration management.

3. 03
   D3 · Claude Code — 3.1 % 6 min read · Level 2

   Its components (rules, skills, commands, agents, memory), session management, slash commands, headless and streaming modes, and the CLAUDE.md and settings.json hierarchy.

4. 04
   D4 · Eval, Testing, and Debugging — 2.6 % 5 min read · Level 2

   Finding errors and recovery strategies, reading traces, and telling whether a problem lives in the integration layer or in the model's output.

5. 05
   D5 · Model Selection and Optimization — 16.8 % 8 min read · Level 2

   LLM fundamentals, SDK, REST and streaming basics, picking the model for each task and managing cost and tokens. With D2 it is half of the exam.

6. 06
   D6 · Prompt and Context Engineering — 11.0 % 7 min read · Level 2

   Context engineering (window, memory, drift, compaction, isolation), prompt engineering (clarity, few-shot, system vs user, constraints) and defensive output handling.

7. 07
   D7 · Security and Safety — 8.1 % 7 min read · Level 2

   Security of AI applications (injection, jailbreaks, PII, leaks), guardrails and safe deployment, hooks as a control, identity, secrets and keys. Few questions, very predictable.

8. 08
   D8 · Tools and MCPs — 10.6 % 9 min read · Level 2

   Implementing tools (schema, descriptions, errors, tool_choice, parallel calls, approval), building MCP servers (tools, resources, prompts, stdio vs HTTP) and when to use each extension point.

9. 09
   Lab 01 — The Messages API in depth 3 min read · Level 3

   A five-line executive summary of a daily price report, generated in under ten seconds: system prompt, parameters, stop reasons and token accounting, measured.

10. 10
    Lab 02 — Tool loop and structured output 3 min read · Level 3

    Turn a scraped product name into valid JSON every time, calling a local catalogue tool when needed: the tool loop by hand, schemas and validation.

11. 11
    Lab 03 — Caching, batches and a cost model 3 min read · Level 3

    Summarize 90 days of reports at the lowest cost: prompt caching for the fixed system prompt, the Batches API for the backlog, and the arithmetic that compares them.

12. 12
    Lab 04 — Production grade: async, errors, logging, versioning 3 min read · Level 3

    Process 200 indicators in under two minutes within the rate limit, survive 429 and 529 without losing work, log every call, and switch prompt or model with evidence.

13. 13
    Lab 05 — An agent with the Agent SDK, hooks and subagents 3 min read · Level 3

    An agent that downloads, transcribes, classifies and files a batch of videos, confined to one folder by hooks and split into subagents.

14. 14
    Lab 06 — An MCP server, a skill and an audit of the kit 3 min read · Level 3

    One MCP server that lets several hosts list, describe and query a synthetic statistical schema read-only, plus a shared skill for the analysis.

15. 15
    Lab 07 — Security: injection, hooks, secrets, PII 3 min read · Level 3

    Attack your own agent with prompt injection, block it with hooks, keep secrets out of the context and redact personal data before it reaches the model.

The project behind it