Senior AI Technical Product Manager - R01563914

Senior Product Manager

Primary Skills

• Prior hands‑on technical or data science experience
• Candidates who have moved from technical or data roles into product, innovation, or applied AI will be particularly strong fit
• Ability to bridge technical depth with product and business thinking

Specialization

• Data Science Advanced: Data Specialist

Job requirements

• Location: PA
• Work type: Remote opportunity with occasional travel to client location
•  

• About the Role
  This is a Technical Product Manager role. You’ll lead a team of engineers, researchers, and data scientists, and you need to be deep enough in the technology to QA AI agent outputs, evaluate RAG pipeline architectures, challenge prompt engineering decisions, review system designs, and guide the team’s technology roadmap. You’re not writing production code, but you’re reading it, understanding it, and making informed calls about whether the technical approach is sound.

• At the same time, you’re the person who builds relationships with internal business teams, understands their pain points, and translates that into the right priorities for a technical team. The difference between this role and a traditional PM is that you’re expected to be in the technical details reviewing architecture decisions, sitting in code reviews, and understanding the tradeoffs between different model choices, retrieval strategies, and deployment approaches.

• You own three things: what the team works on (prioritization), how well it’s done (quality), and whether it matters (stakeholder alignment). You lead planning and serve as the quality gate every piece of work passes through your technical and product review before it ships or gets shared with stakeholders.
  As the team’s solutions scale, this role will also evolve into LLM cost optimization as a budget line, navigating AI governance and responsible AI practices, and guiding the technology roadmap as the AI landscape shifts.

• Responsibilities
  Product & Stakeholder Management
  • Set priorities and decide what the team works on across both AI Enablement and AI Experiments tracks
  • Serve as the primary liaison with internal business teams understand their workflows, gather requirements, and translate pain points into actionable technical work
  • Collaborate with product teams to ensure exploration work aligns with the broader product direction
  • Lead the team’s operating rhythm stand-ups, demos, planning sessions, and progress readouts to leadership and stakeholders
  • Make resource allocation decisions across workstreams move people where they’re needed most based on shifting priorities
  • Decide when to kill work that isn’t delivering results you’re comfortable shutting down experiments and reprioritizing without hesitation
  • Communicate progress clearly to leadership and cross-functional partners what shipped, what we learned, what’s next, and what needs their attention
  • Identify new internal processes where AI could meaningfully reduce effort, and build the business case to expand the team’s scope
  Technical Leadership
  • Guide the team’s technology roadmap make informed decisions about model selection, infrastructure choices, build-vs-buy tradeoffs, and when to adopt new tools or frameworks. You stay current on how the AI landscape is evolving and what that means for the team’s technical strategy.
  • QA all AI outputs before they move forward review agent responses for quality, accuracy, and edge case handling. You need to understand why an agent produced a given output and whether the underlying prompt, retrieval, or orchestration logic is sound.
  • Review and guide architecture decisions evaluate RAG pipeline designs, agent orchestration patterns, data flow architectures, and integration approaches. You’re the person who asks “why did we choose this embedding model?” or “what happens when the retrieval layer returns irrelevant context?” and expects a rigorous answer.
  • Evaluate experiment designs for technical rigor when the team tests a new model or prompting approach, you review the evaluation methodology, success criteria, and whether the comparison is fair before the experiment runs, not after.
  • Participate in technical design reviews and code reviews not to approve every line of code, but to stay close enough to the implementation that you can catch architectural issues, scalability concerns, and technical debt before they compound.
  • Manage and optimize AI infrastructure spend track LLM costs, token usage patterns, and vendor contracts. You understand the cost implications of different model choices, caching strategies, and prompt