Founding cohort·$180 → $150 for the first cohort.·Tuition returns to list once cohort 1 closes.
Applied LLMs for Builders
A working engineer's path through modern LLM application development.
A working engineer's path through modern LLM application development — prompting, evaluation, retrieval, agents, and shipping to production.
For working engineers and technical leads building with AI today.
You will be able to
do the work.
- Build and evaluate retrieval-augmented systems end-to-end
- Design and run rigorous prompt and model evaluations
- Reason about cost, latency, and quality trade-offs across providers
- Ship a production LLM feature with monitoring and guardrails
What we cover.
Prerequisites
- Comfortable in Python or TypeScript
- Have shipped at least one production service
- Familiar with HTTP APIs and basic data work
- 01
Module 1 — Prompting as engineering
Treating prompts as code: versioning, testing, and the failure modes you only see at scale.
- 02
Module 2 — Evaluation that holds up
Building eval suites that survive a model swap. Golden sets, judges, regressions, drift.
- 03
Module 3 — Retrieval and grounding
Chunking, embeddings, hybrid search, reranking. What works, what's cargo-culted, and how to tell.
- 04
Module 4 — Agents and tools
When to reach for an agent loop and when not to. Tool design, structured outputs, recovery.
- 05
Module 5 — Production
Latency, cost, observability, guardrails, and a capstone shipped to real users.
Reserve a place.
We open enrollment a few weeks before each cohort starts. Tell us about your work — we'll write to you first when a place opens.
Founding cohort. Small group, two evenings per week + a capstone weekend. Tuition returns to list once cohort 1 closes.