means
ends
dspy prompt-engineering claude-code ai-tools optimization
How we built a prompt optimization system that makes Claude Code's agents measurably better at their jobs. Bootstrap algorithms, semantic equivalence groups, and the problem of defining 'better' when the judge is also an AI.
The inaugural post. How this blog was built by talking to Claude, what three tiers of web complexity look like, and why radical transparency about the process is the point.
Ask about my projects
Hi! I can answer questions about Ashita's projects, the tech behind them, or how this blog was built. What would you like to know?