One-Day Intensive
Context Engineering
10.23.26 | 10:00 AM - 4:00 PM PDT 6 hours in one focused day
Engineer the full context lifecycle, build agents that remember, retrieve, and defend.
with Liz Howard
Overview
“Context” is the whole game, and most people fill it with noise. This hands-on Python intensive walks the full lifecycle of model context: what goes in, why, how to budget it, how to persist it, and how to defend it. By 4pm you'll have built, from scratch, a token-budget tracker, a FAISS vector store with RAG retrieval, multi-stage summarization pipelines, a sliding-window context manager, a 5-tier hierarchical memory system, and a red-team suite that throws 10+ jailbreak and injection attacks at your own agents, plus a Constitutional AI critique-revise loop to defend them. You'll leave knowing what “context” actually means, and how to engineer it.
What You'll Master
- Engineer the full context lifecycle, what goes in, why, budget, persist, defend
- Build a FAISS vector store and RAG retrieval from scratch
- Compress long docs with summarization pipelines and sliding-window management
- Give agents a 5-tier hierarchical memory that persists
- Red-team your context, beat 10+ jailbreak/injection attacks, add Constitutional AI
Who this is for & what you'll need
Pro engineering
Built for working engineers.
Strong computer or subscription
Use your own capable machine, or a paid hosted option below.
On your own machine
Install LocalLM and run open-source LLMs locally for free (for those with the hardware).
Or use the hosted option
Or use the OpenAI API / ChatGPT Plus, with Google Colab (Oobabooga) for the open-model work.
Where This Class Leads
What you leave able to do
- re architect an agent as a stateless reducer
- reconcile a failing retrieval by reranking and rewriting
- reconcile a working set with the context window it must fit
- reconcile an agents run with a tool that failed
- select among agent architectures
- synthesise a memory hierarchy for an agent
- synthesise an agent that carries notes across its own runs
How you show it. Paired runs of the same multi-step task with the note store kept and cleared, where the cleared run repeats a step the noted run skips, plus the note text the agent wrote and the later turn that cites it.
It sits on 4 paths
Build production AI systems from prompt to deployment.4 of 12 after AI Alignment · before Claude Model Context Protocol Replace repetitive business operations with AI (a mini-MBA in AI).4 of 4 after Custom GPTs Founder with an AI-first company.5 of 7 after Automate Your Email: No Code AI Automation · before Intro to Agents Build strange systems that change how people think.6 of 8 after Automate Your Email: No Code AI Automation · before The Dark Arts (Red Team)About 120 hours of practice sit behind this one. See what your hours buy.
Join Us
One-Day Intensive
- Full day intensive workshop, community access, advanced AI integration skills.
- Your seat in the class
- Lifetime access to materials
- Sponsors 1 scholarship seat for someone else
The minimum is $200, it keeps out spam. Need to pay less? Apply for a scholarship, no one is turned away for lack of funds.
Secure payment via Stripe
No sponsored scholarship seats are open in this class yet. Not enrolling, but want to put someone through this class?
Instructor
Want to schedule this class for your team?
Contact us: [email protected]