Odyssey · San Francisco
Product & ProgramOperationsFullTimeposted 8w ago
Odyssey is an AI lab pioneering general world models: causal, multimodal systems that learn to predict and interact with the world over long horizons. This foundational technology promises to revolutionize robotics, science, healthcare, education, gaming, defense, and beyond.
Odyssey’s founders previously pioneered the most complex application of physical AI: self-driving cars. They’ve now brought together a world-class research team from DeepMind, Tesla, Waymo, Meta, Apple, and Wayve, who have made significant contributions to language models (DeepMind Gemini), video models (DeepMind Veo), world models (Wayve GAIA), and autonomous systems (Tesla FSD).
Odyssey has raised significant venture capital from GV, Amazon, AMD, EQT, NVIDIA, Natural Capital, In-Q-Tel, Elad Gil, Jeff Dean, Guillermo Rauch, Garry Tan, Kyle Vogt, and researchers from OpenAI, DeepMind, MSL, Recursive, and Thinking Machines.
The Data Program Manager is the connective tissue between research, vendors and the data flywheel. You are responsible for understanding our data needs by being deep in the frontier research space for multi-model methodologies and arxiv papers. You are responsible for presenting a credible data roadmap to the research, applied ML, and leadership and keeping stakeholders abreast of the pipeline and plans. You also own inbound requirements and turn those into a backlog the data-ops team can execute against.
The right person thinks in processes, has strong opinions, and can hold both people (internal and external) and plans accountable.
Run data plans with loose requirements from Research. Turn fuzzy, disparate data requests into scoped, sequenced acquisition plans with quality bars, timelines, and clear data parameters.
Manage a multitude of vendors across verticals and data types, maintaining relationships and negotiating commercial terms favorable to Odyssey.
Manage the human teams (experience with contractors a big plus) that acquire proprietary data and establish and enforce quality measures.
Stand up and own the weekly data-needs review and forecast. Force-prioritize when necessary and surface the trade-offs to leadership.
Embed in research planning. Flag data risks four to six weeks ahead of when they would hit a model run.
Define and track quality and throughput metrics across active data programs. Redline and QA batches that slip, and close the loop with vendors.
Partner with Research on data requirements while also running independently on frontier data mix delivery.
Drive program retrospectives, what data moved which evals, at what cost, and what we should repeat.
5+ years of program or product management at a technical company, with at least two of those years in an AI, ML, or data-intensive context.
You can read ML papers without flinching; you don't need to derive the maths, but you understand what the research aims to prove and what data would support it.
Working knowledge of SQL and basic data analysis. You can pull your own numbers when needed.
Real experience negotiating with external vendors and driving down costs.
You can run a tight stand-up, keep clean docs, and finish things.
Comfortable saying "no" to a researcher; better at saying "yes, by week six, here's the plan."
Track record of driving cross-functional programs at speed in environments where priorities change weekly.
Experience as a TPM or program manager at a frontier research / applied research lab.
Background in data labelling, human data collection, or RLHF programs at scale.
Familiarity with AI evaluation methodology or rubric design.
Prior experience as the first X hire at an early-stage company.