Following the boundary between autonomy and accountability.
A living page about the questions, systems, and practices receiving my attention right now.
Autonomous workflows are getting easier to assemble. The more interesting product problem is deciding what they should be allowed to do, how we know they are doing it well, and where human judgment creates more value than another model call.
Agentic product patterns
Where autonomy creates leverage, where it creates hidden risk, and how an interface should expose the difference.
Evaluation as infrastructure
Turning product judgment into living test sets, production signals, and feedback loops that a whole team can use.
Human–agent handoffs
Designing context, confidence, and accountability across the point where an automated workflow returns control to a person.
Technical range for PMs
Staying close enough to code, models, and architecture to make better product decisions without confusing depth with ownership theatre.
Frame the decision. Build the smallest real system. Watch where reality disagrees.
- 01 Frame
- 02 Build
- 03 Observe
- 04 Revise
If one of these questions overlaps with yours, that is probably worth a conversation.
Send a note