Overview As Technical Product Manager for Software and Agents, you own the core platform layers and agent-driven workflows. You translate research needs into a coherent product by aligning engineers, ML researchers, and scientists around a shared roadmap. You govern what belongs in the core platform, guide API and service design, and ensure production readiness. This hands-on role shapes how researchers set goals, how AI agents plan work, and how experimental results improve future decisions. You will work onsite in Brooklyn, helping scale a cutting-edge AI and science platform toward real-world impact. ResponsibilitiesOwn roadmap for shared software, data, and infrastructure supporting agents, models, and lab systemsDefine core vs edge platform boundaries and build-versus-adopt decisionsShape core services and APIs for workflow orchestration, experiments, and data coordinationDefine data capture, structuring, versioning for research and literature dataSet reliability, performance, observability, and production readiness expectationsDefine agent capabilities, tools, and human review requirementsDesign user interfaces for goal setting, plan review, and result tracingEstablish metrics and acceptance criteria tied to lab and customer outcomesGuide model deployment decisions (build, fine-tune, adopt) and the learning loop with experimentsCoordinate with robotics and autonomy PM to define platform-lab system interfacesCommunicate product direction to engineering, research, and leadership Key requirements5-7 years in product managementat least 1 year on AI agents or ML-based productstechnical credibility with ability to read code and reason about architecturesability to discuss feasibility and tradeoffs with engineers/researchersstrong judgment on AI agent limits, tool reliability, grounding, and oversightexperience linking model/agent metrics to customer or operational outcomesability to integrate agents, models, UX, and platform foundations into a cohesive productproven ability to align engineering, research, science, and design teamsstrong first-principles thinking in ambiguous environmentsstrong cross-functional collaborationclear decision-making and prioritizationcommunication of technical concepts to non-technical stakeholdersread code and build lightweight prototypesreason about agent architectures, APIs, data flows, and integrationsknowledge of experiment design and data governance
Not specified in the original listing.
Not specified in the original listing.
