Level 4
AI Professional
Take AI to production — safety, cost, latency and reliability at real-world scale.
- 124 min→
Guardrails and Safety in Production
Ship AI that strangers use without getting burned: defend against bad inputs, unsafe outputs, prompt injection and leaked data — the difference between a demo and something you'd put your name on.
- 222 min→
Cost and Latency at Scale
Make AI features fast and affordable enough to run for real users: the levers on token cost and response time, and the engineering patterns — caching, routing, streaming — that tame both.
- 320 min→
Observability: Seeing What Your AI Does
You can't improve what you can't see. Learn to instrument an AI system so you know what it's doing in production — logging, tracing, monitoring quality, and closing the feedback loop.
- 422 min→
Fine-Tuning and Customization
Know the real toolbox for adapting a model to your needs — prompting, RAG, and fine-tuning — and, crucially, when each is the right (and wrong) answer. Most teams reach for fine-tuning too early.