Data Program Manager - Frontier AI Lab Equity

Palo Alto (Hybrid) | Up to $300k plus Bonus & Equity

Our client is building general-purpose, causal, multimodal AI systems that learn to predict and interact with the world in real time. Think of it as teaching AI to understand physics, motion, and environments the way humans do, then letting you step inside and interact with it. Their latest model holds the highest physics score among evaluated world models and runs in real time.

This isn't incremental. It's a new medium.

The models are only as good as the data they learn from, and this role owns the strategy, sourcing, quality, and end-to-end delivery of that pipeline. You'll be the connective tissue between research, vendors, and the data flywheel.

Your first and most important job: taking full ownership of the vendor ecosystem. Screening new vendors, negotiating commercials, reviewing SOWs and contracts, and holding partners to quality bars and timelines. From there, you'll take loose requirements from researchers and turn them into scoped acquisition plans, embed them in research planning, flag data risks weeks before they hit a model run, and drive retrospectives on what data moved which evals at what cost.

This is a role for someone who thinks in process, holds people and plans accountable, and isn't afraid to say "no" to a researcher or, better yet, "yes, by week six, here's the plan."

• Own vendor management end-to-end screening, selection, commercial negotiation, SOW and contract review, and ongoing performance management across data types and verticals
• Turn fuzzy data requests into sequenced acquisition plans with quality bars, timelines, and clear parameters
• Stand up and own the weekly data-needs review, force-prioritise when necessary and surface trade-offs to leadership
• Embed in research planning and flag data risks 4-6 weeks ahead of model runs
• Define and track quality and throughput metrics; automate quality reviews where possible, redline batches that slip, and close the loop
• Drive program retrospectives: what worked, what didn't, what to repeat
• Manage human data collection teams (contractor experience is a big plus)

• 5+ years in program or product management at a technical company, with at least 2 in an AI/ML or data-intensive context
• Serious vendor chops: real experience screening vendors, negotiating terms, reviewing SOWs and contracts, and driving costs down.
• With deep familiarity with the AI model lifecycle and experience working directly with frontier AI researchers, you understand how data mixtures shift across pre-training, fine-tuning, and evaluation phases
• You can read ML papers without flinching, you don't need to derive the math, but you understand what the research is proving and what data would support it
• Working knowledge of SQL and basic data analysis; comfort using AI coding agents to analyse datasets and automate quality workflows
• You run a tight standup, keep clean docs, and finish things
• Track record of driving cross-functional programs at speed in environments where priorities shift weekly
• TPM or program management at a frontier AI lab
• Background in data labelling, human-data collection, or RLHF at scale
• Familiarity with AI evaluation methodology or rubric design
• Prior experience as a first hire at an early-stage company

This is a chance to shape the data strategy at one of the most exciting AI labs in the world - where the models are learning to simulate reality itself, and the data you source directly moves the frontier.