Lead AI Product Manager Retirement & Wealth Domain

Lead AI Product Manager with Retirement & Wealth Domain

Boston, MA or Windsor, CT

We need candidate to work onsite from Day 1 (Onsite Hybrid)

• Discovery & Specification
• Execution & Delivery
• Stakeholder Alignment

• 8+ years of product management experience, with at least 4 years in AI/ML product roles at a technology company, fintech, or financial services firm.
• Demonstrated track record of shipping AI-powered products to production—owning the full lifecycle from discovery through measurable adoption.
• Lead or principal-level experience: defined product strategy and roadmap independently, not just executed against someone else’s vision.
• Prior ownership of products in a regulated environment (financial services, healthcare, or similar); experience navigating compliance and legal review as part of the standard product process.
• Experience influencing VP-and-above stakeholders without direct authority.

Evaluated rigorously. Candidates should expect to demonstrate these in the interview process, not just claim them on a resume.

• LLM product experience: shipped at least one production feature using large language models (OpenAI GPT-4o, Anthropic Claude, Google Gemini, or equivalent); understands prompt engineering, system prompt design, context window management, and structured output extraction.
• RAG architecture fluency: can evaluate the quality of a RAG pipeline—chunking strategy, embedding model selection, retrieval precision/recall trade-offs, re‑ranking logic, and hallucination mitigation. Does not need to implement but must be able to interrogate.
• Agentic AI product design: has designed or shipped features using agentic workflows (tool use, multi‑step reasoning, agent orchestration via Lang Chain, Lang Graph, Vertex AI Agent Builder, Copilot Studio, or equivalent); understands where agents fail and how those failures affect fiduciary use cases specifically.
• Model evaluation and metrics: can define evaluation frameworks for AI outputs; understands precision/recall, ROC‑AUC, hallucination rates, and task‑specific quality metrics; able to review an LLM eval suite and assess whether it covers the right failure modes for a retirement context.
• Data fluency: comfortable interrogating SQL, reviewing data pipeline design, and forming hypotheses from participant behavioral data without requiring a data analyst to translate.
• AI tooling in practice: uses AI coding assistants (Git Hub Copilot, Claude Code, Cursor, or equivalent) and agentic tools daily—this team builds with these tools, not about them.
• API and system awareness: can read a technical architecture diagram, understand latency/reliability constraints, and write specs that account for engineering realities including model serving costs and token limits.
• Experimentation: A/B test design, cohort analysis, statistical significance, and shadow deployment patterns for AI features in production.

• Defined Contribution Plans: 401(k), 403(b), 457 mechanics; contribution limits and catch‑up provisions; employer match and vesting design; recordkeeper/TPA/plan sponsor ecosystem; QDIA rules; plan document fundamentals.
• ERISA & Fiduciary Standards: ERISA prudence and loyalty requirements; functional fiduciary standard and prohibited transactions; how AI‑generated outputs must be structured to support — not replace — fiduciary decision‑making; DOL guidance on AI use in retirement plan contexts.
• 2026 Regulatory Landscape: SECURE 2.0 provisions (auto‑enrollment, RMD changes, catch‑up rules); the April 2026 interagency model risk management guidance superseding SR 11‑7—including its principles‑based approach to materiality tiering and proportional controls for AI and agentic systems; evolving DOL fiduciary rule.
• Participant Behavior & Retirement Readiness:
  Behavioral finance drivers of savings inertia; retirement income adequacy frameworks; auto‑enrollment and escalation research; decumulation and guaranteed income strategies (relevant to SECURE 2.0 lifetime income provisions).
• Investment Products:
  Target‑date fund construction and glide paths; managed account structures…