Self-improving agents are an attractive idea
The governance implications are uncomfortable
Understanding is the new moat
Do you understand what you produced, and can you defend it when it's wrong?
The lawyer who builds: governing AI from the inside
A lawyer whose raw material is judgment, who builds enough to make that judgment real, and who earns the right to govern a thing because he understands it from the inside
Your observability stack is your AI audit trail
whether you planned it that way or not
Legal Tech Just Split in Two
Mind the Money
When software goes headless, who's watching the agents?
Enterprise compliance was built for human-operated interfaces.
AI agents could dissolve the friction that keeps justice expensive
It restructures who holds legal power, where authority comes from, and when law intervenes.
Harness Engineering: Why Better Models Aren't Enough to Ship AI Agents
That gap is "harness engineering," and it's where the real work of building production-grade agents happens
