Research Product Manager (RPM) (US)

What you'll do

• Co-design model capabilities by identifying gaps in frontier models, running headroom analyses, and defining areas of improvement alongside researchers at top AI labs
• Run experiments and benchmarks using internal simulation infrastructure and open-source tooling (e.g., Oddish) to evaluate model performance and hillclimb on specific verticals
• Translate customer requirements into action, take vague asks from partner labs, break them into research questions, and organize internal teams to hit aggressive delivery timelines
• Design and manage data curation pipelines, recruit and coordinate external domain experts, or work with research engineers to build synthetic data generation pipelines
• Perform literature reviews, stay current with SOTA research, read papers, and implement insights into practical training data and evaluation frameworks
• Own the full project lifecycle from research scoping through weekly data delivery, incorporating customer feedback and iterating rapidly over 1-3 month project cycles
• Bridge research and operations, communicate technical findings to non-technical contributors and translate user/enterprise pain points into evaluation and training data

What you need

• We need someone with 0-5 years of experience in AI/ML research or technical product management who is deeply technical, research-minded, and comfortable navigating ambiguity at a fast-paced early-stage startup. You should be fluent in LLM evaluations, benchmarks, and coding agents, with a track record of shipping technical projects end-to-end. Bonus points if you have founder experience or have worked at AI data companies or frontier labs like Anthropic, OpenAI, or similar.

Frequently Asked Questions

Related Roles