Machine Learning Engineer

Machine Learning Research EngineerSan Francisco, CA | On-site | Relocation Support AvailableWhat if the next breakthrough in AI isn't a bigger language model?One of the world's most ambitious AI research teams believes we've been asking the wrong question.Instead of building larger language models, they're developing a completely new class of foundation model capable of understanding cause and effect—AI that can predict what will happen next and determine the actions required to change the outcome.Their starting point isn't text.It's the physical world.Using one of the largest collections of real-world observational data available, this team is building large-scale foundation models that learn directly from physical systems, with the long-term goal of advancing a new generation of general intelligence.If you're motivated by solving problems that don't yet have published answers, and want to help shape a research direction rather than optimise an existing one, this is an exceptional opportunity.The Posiiton.You'll work alongside a small team of world-class researchers and engineers building large-scale multimodal foundation models from the ground up.This is a hands-on research engineering position spanning the entire machine learning stack, from data infrastructure and distributed training through to model architecture, evaluation, experimentation and production research systems. You'll be expected to challenge assumptions, move quickly, and contribute directly to the core research agenda.What You'll Be DoingTrain large-scale foundation models from scratch.Design and implement novel model architectures and training approaches.Build distributed training infrastructure across hundreds of GPUs.Develop petabyte-scale multimodal data pipelines.Design evaluation frameworks and analyse experimental results.Collaborate closely with researchers to rapidly iterate on newideas.Contribute across modelling, infrastructure, experimentation and systems engineering.We're Looking ForYou'll likely bring experience in several of the following:Training large-scale foundation models rather than simply fine-tuning existing ones.Distributed training using PyTorch, FSDP or similar frameworks.Building large-scale ML infrastructure and data pipelines.Strong machine learning fundamentals.Experience in one or more of: Large Language Models Computer Vision Robotics Sensor Fusion Physics-informed Machine Learning Scientific AI Multimodal Learning Comfortable working across research and engineering. The EnvironmentThis is not a large corporate research lab.You'll join a highly funded early-stage company with fewer than 15 people today, scaling rapidly over the next year.The team is intentionally small, highly collaborative and deeply technical.Everyone writes code.Everyone contributes to research.Everyone is expected to move quickly.The role is based five days a week in San Francisco, with relocation support available.Why This Opportunity?Most AI companies are trying to scale today's models.This team is attempting to build the next paradigm.If you're excited by frontier research, enjoy solving problems that don't yet have established solutions, and want to work with a small group tackling one of AI's hardest challenges, we'd love to hear from you.