Somewhere between curiosity and caffeine โ, I started teaching computers to do interesting things.
What began as small scripts slowly grew into systems โ agents that explore the web, models that try to understand language, and tools that help machines reason about messy real-world data.
Lately I've been spending a lot of time experimenting with AI systems: building agents, designing pipelines, and trying to understand how intelligent software should interact with the world.
Some experiments work. Some experiments break spectacularly. Both are useful.
- โ Coffee is still the most reliable debugging tool
- ๐ง I enjoy turning complex problems into systems that make sense
- ๐ I like building tools that build other tools
- ๐ค Currently fascinated by autonomous agents and AI systems that can explore, reason, and learn
A collection of experiments around:
- AI systems and agents
- tools for working with data and language
- infrastructure that makes intelligent software actually run
Some projects are prototypes. Some are half-finished ideas. Some are things that turned out better than expected.
All of them are part of the same curiosity: how far we can push intelligent software.
If you're building interesting things in AI, systems, or anything slightly strange โ I'd love to see it ๐