Reading up on RunLLM
2 deep · digging since jan 18
- Agents can't choose between structure and flexibility
The article argues that employing pure code or pure markdown to specify AI agent behavior is a failure mode, and advocates for a hybrid approach.
- Data is your only moat
The article argues that data collected from user adoption is the only sustainable moat for AI applications, with easy-to-adopt, easy-to-solve use cases being value traps for startups.