Using AI Effectively In Large and Complex Codebases

AI tools work great on toy projects but struggle at scale. This talk covers how to think about LLMs in the context of large, complex codebases: what they're genuinely good at, where they break down, and how to structure your work to get output that is consistently better.

We'll cover practical techniques for writing effective prompts, managing context windows, and leveraging skills, plugins, and tools to keep AI productive across a real codebase. You'll also learn what to address in your codebase before you try to accelerate, so AI becomes a force multiplier rather than a source of rework.

It's on the 5th floor in the community room. Turn right after leaving the elevators.