Week 1
Foundations
Ship a deployed AI agent on day one. Then make it useful.
- #01Hello, AgentFree
- #02Give it memoryFree
- #03Stream the responseFree
- #04First tool callFree
- #05Persist to PostgresFree
Curriculum
Six tracks. One workshop.
12 weeks · 60 sprints · 1 capstone per track. Pick one to focus on. Branch into others as you grow.
12-week track
AI Engineering
Now enrollingEnter track
Capstone
Build a production-grade RAG assistant with evals and cost guardrails — deployed and defensible in interviews.
Week 2
Tools & Plans
Multi-step reasoning: pick the right tool, then chain them.
- #06Tool registry pattern
- #07Routing — which tool when
- #08Multi-step plans
- #09Cost tracking
- #10Prompt caching
Week 3
Embeddings & Search
Make the agent retrieve relevant context instead of guessing.
- #11Embeddings 101
- #12Vector store: pgvector
- #13Naive RAG
- #14Chunking strategies
- #15RAG over a real doc set
Week 4
Production RAG
Naive RAG breaks at scale. Fix retrieval, citations, and failure modes.
- #16Reranking
- #17Hybrid search
- #18Source-grounded answers
- #19Empty-result handling
- #20RAG evals
Week 5
Agents
Beyond single calls — reasoning loops, planning, self-correction.
- #21ReAct loop
- #22Multi-agent: planner + worker
- #23Self-correction
- #24Long-horizon decomposition
- #25State machines for agents
Week 6
Production
Make it real: real users, real auth, real observability.
- #26Deploy to Vercel
- #27Add auth (Clerk)
- #28Multi-tenant isolation
- #29Cost guardrails
- #30Structured logging + tracing
Week 7
Quality
If you can't measure it, you can't ship it.
- #31Eval harness from scratch
- #32LLM-as-judge
- #33Regression catching
- #34A/B testing models
- #35Red-team prompts
Week 8
UX
The model is the easy part. The UX is the product.
- #36Streaming UX patterns
- #37Structured outputs
- #38Show the work
- #39Interruptible generation
- #40Optimistic UI for agents
Week 9
Multimodal
Beyond text — images, documents, audio, code, browsers.
- #41Vision input
- #42PDF understanding
- #43Audio: STT + TTS
- #44Computer-use agent
- #45Code-writing agent
Week 10
Scale
What breaks at 10k users? Fix it before they show up.
- #46Prompt caching at scale
- #47Request batching
- #48Latency budgets
- #49Provider failover
- #50Edge inference
Week 11
Hardening
Production AI = security AI. Adversaries are now your users.
- #51Prompt injection defense
- #52Jailbreak resistance
- #53PII redaction pipeline
- #54Rate limiting + abuse
- #55SOC 2 readiness
Week 12
Capstone
One real project. Ship it. Defend it. Add it to your portfolio.
- #56Capstone: design
- #57Capstone: v1 ship
- #58Capstone: add evals
- #59Capstone: production hardening
- #60Capstone: demo day + handoff