What you’d own
- own the production improvement loop for agent behavior and outcomes
- instrument agent workflows, tool use, decisions, and failures
- define quality standards, evaluation datasets, and regression coverage
- investigate agent failures across context, tools, and workflow design
- ship improvements to prompting, context, decision logic, and tool use
- build backend services, APIs, data models, and feedback pipelines
- run experiments, production replays, and staged rollouts
- partner with GTM teams to define customer and business success
From the posting
See the full posting for context and requirements
This is a summary; check the source posting for complete, current requirements.