This list contains only the countries for which job offers have been published in the selected language (e.g., in the French version, only job offers written in French are displayed, and in the English version, only those in English).
Luma's mission is to build unified general intelligence that can generate, understand, and operate in the physical world. We believe that multimodality is critical for intelligence. To go beyond language models and build more aware, capable and useful systems, the next step function change will come from vision. So, we are working on training and scaling up multimodal foundation models for systems that can see and understand, show and explain, and eventually interact with our world to effect change. You'll sit between Luma's frontier research efforts and our core products — Canvas, Agents, and the model platform — ensuring that what researchers build is informed by what customers need, and that what products ship takes full advantage of what research makes possible.
Job Responsibility
Own the product strategy for translating Luma's frontier research capabilities into shippable features across Canvas, Agents, and the model platform
Work directly within the research team to shape priorities, ensuring model development and evaluation are informed by real enterprise needs in marketing, advertising, and entertainment
Establish and operate the feedback loops between research, product, go-to-market, and forward-deployed teams — so that market signals flow into research prioritization and research breakthroughs flow into product planning
Interpret evaluation results and research findings to identify which emerging capabilities have the highest product and commercial leverage — and make bets accordingly
Build prototypes yourself to validate product ideas before committing engineering resources — using the models directly to test hypotheses about what's possible and what customers will value
Define the product implications of research advances in multimodal understanding, generation, tool use, and agent capabilities — translating "the model can now do X" into "customers can now solve Y"
Partner with the Agent PM and Enterprise PM to ensure research priorities align with the capabilities that enterprise products and customer deployments actually need
Drive the cadence and process for research-to-product handoffs — defining what "ready to ship" means, managing the transition from research prototype to production feature, and owning the quality bar
Requirements
Principal or Staff-level product management experience (L6 seniority or above) with a track record of bridging pure research and ambitious product development at an AI lab or applied AI company
You are genuinely comfortable in a research environment — you can read evaluations, understand what they mean, ask the right questions, and help researchers see the product implications of their work
You've shipped products that originated in research — not just optimized existing products, but taken emergent capabilities and turned them into new product surfaces
You can build prototypes yourself — you have the technical depth to work directly with models, write prompts, test hypotheses, and validate ideas before involving a full engineering team
You operate with extremely high agency and a strong self-starter mentality — research environments are inherently ambiguous and this role has no existing playbook
You have a general manager mindset — you can evaluate research directions not just for technical novelty but for commercial viability and customer impact
You bring a mix of large-company and earlier-stage experience — you understand both the pace of a research lab and the urgency of a startup building a commercial business
You can influence without authority — researchers don't report to you, but your input on priorities needs to carry weight because it's well-informed and well-reasoned
Nice to have
Experience as a Research PM, Applied Research PM, or Labs PM at a frontier AI company (Anthropic, OpenAI, Google DeepMind, Meta AI, Cohere, Mistral)
Background in multimodal AI — you understand the specific opportunities and challenges of models that work across vision, language, and generation
Python and SQL proficiency — you can work in the tools researchers use, not just the tools PMs use
Experience defining evaluation frameworks and quality bars for shipping research capabilities into production products
Founder-type experience or background — you've operated at the intersection of technical ambiguity and commercial urgency before
Domain exposure in marketing, advertising, entertainment, or related verticals where multimodal capabilities have the most immediate commercial application