Events
Two formats where you don't learn alone: an intensive with a mentor that ends in a demo day, or a weekend competition. No dates yet — waitlist signups decide which one launches first.
We'll announce dates and prices once the first group fills up. Join the waitlist and you'll hear first.
Two weeks with a mentor: bring a task from your own job and leave with an assistant that handles it reliably — with evals to prove it.
Duration: 2 weeks
Outcome: A deployed LLM assistant for your task, backed by a set of evals
For: Non-coders — ops, marketing, managers — who use an AI chat every day and are tired of regenerating answers
Three weeks, three automations of a real process. Agent loops, TDD prompts and pipelines that never fail silently.
Duration: 3 weeks
Outcome: 3 working automations of a real process
For: Developers who already use coding agents and want to move from one-off prompts to pipelines that run on their own
One week to build an MVP through agents — from spec to deploy, with subagents, MCP and hooks.
Duration: 1 week
Outcome: An MVP built by agents, from spec to deploy
For: Daily Claude Code users who still drive it like a chat box
48 hours online: break other people's LLM systems and defend your own. Attack, defense, one leaderboard.
Duration: 48 hours, online
Outcome: A spot on the attack-and-defense leaderboard for LLM systems
For: Prompt engineers and LLM product builders who need to know how their system will be attacked
One weekend, one brief for everyone: turn it into a spec and build an MVP from it. Judged by a jury.
Duration: A weekend
Outcome: A team MVP built from the brief, scored by a jury
For: Developers and PMs who want to see how fast they can ship a product with agents
A weekend on real tasks from partner companies. Partner enrollment is open — tasks are announced with the date.
Duration: A weekend
Outcome: A solution to a real company problem, plus sponsor prizes
For: Practitioners who want to solve a real business problem rather than a training one