Luma AI jobs, judged for you, free

What they do

Luma builds multimodal AI models and apps that make video and images, such as its Ray video model. Creators, agencies, brands, developers, and companies use its models in a creative app or through its API. Luma also builds world models for the physical world. It started in 2021 and says it has more than 30 million users.

What they value

• Build AI that can understand and act in the physical world.
• Unite research, product, and sales into one engine.
• Trust people to manage their own time.

Where they hire

Of 28 Postings, ten name Redwood City and ten name London. Others name New York, Seattle, Munich, Los Angeles, and Riyadh, and many research roles are also open to remote work in the US, the UK, or the EU. Twenty-three Postings are hybrid and five are remote. No Posting mentions visa sponsorship.

Teams and roles

Research and AI has 16 of the 28 Postings.

• Research and AI: four roles posted in several places, for performance optimization, reinforcement learning infrastructure, training infrastructure, and inference. Others work on world models, simulation, AI agents, and data infrastructure.
• Go-to-market: four sales executives, three forward deployed creatives, two senior customer success managers, and a delivery lead.
• An editor for motion graphics in Los Angeles, and a general application.

What they look for

Fourteen of the 16 research Postings ask for PyTorch, and many ask for training at scale on GPU clusters. Performance roles need CUDA, Triton, and profiling tools. Reinforcement learning roles need experience post-training large language models. Inference engineers need Python, vLLM or SGLang, Docker, and Kubernetes. Sales executives need 5+ years of full-cycle sales, and customer success managers 5 to 10+ years. Forward deployed creatives need senior client work and generative AI creative tools. Some sales and creative roles need French or German.