Lightning Lessons

Advanced Context Engineering

Hosted by Dexter Horthy and Meri Nova

Sat, Sep 6, 2025

7:00 PM UTC (45 minutes)

Virtual (Zoom)

Free to join

66 students

Invite your network

Go deeper with a course

Agent Engineering Bootcamp
Meri Nova and Hai Nghiem
View syllabus

What you'll learn

How to use AI with larger codebases

Challenges in AI-generated coding, especially for complex problems and brownfield codebases

How to optimize AI Coding Agents

Techniques to make AI coding agents work well for hard problems in large codebases

Practical examples from the real world

Examples of using AI to make changes to 300k-line Rust codebases. Takeaways and next steps.

Why this topic matters

Our guest speaker, Dex, will dive into the evolution of context engineering for building reliable LLM apps, rooted in the “12 Factor Agents” manifesto. Learn from his experience generating 2,000-line PRs, fixing issues in a 300k-line Rust codebase with techniques like: Intentional compaction, Sub-agents research-plan worklfow To ship product-quality code

You'll learn from

Dexter Horthy

Example: VP at Top Company (ex-Role at Previous Company)

I'm Dex, and I'm hacking on coding agent maxxing at HumanLayer, building the best way to accelerate AI-native dev for complex tasks in brownfield and legacy codebases.


Before this I was working on AI Agents that managed SQL warehouses, and did a long stint at replicated.com helping the worlds best software teams deliver Kubernetes apps into customer environments.


I've been coding since 17, when I built tools for NASA researchers to navigate the south pole of the moon. Enjoyer of tacos and burpees (not necessarily in that order)

Meri Nova

Founder @Break Into Data.

Host of the Agent Engineering Bootcamp.

Founder at Break Into Data.


Linkedin - Meri Nova

Previously at

NASA
Y Combinator

Sign up to join this lesson

By continuing, you agree to Maven's Terms and Privacy Policy.