Pantograph PBC
Research Engineer
San Francisco · On-site · Full-time · Posted Aug 11, 2026
- Education
- No specific credential required; the posting prioritizes what candidates have built.
About the job
Train and improve general models using large-scale video and other diverse datasets, while building the research and engineering systems needed to run experiments at scale.
What you'll do
- Implement research ideas and scale experiments across large GPU clusters.
- Build systems and infrastructure for model training, data processing, evaluation, and rapid iteration.
- Investigate experiment results and system behavior to understand and resolve problems.
What they ask for
- Experience training models across large GPU clusters and comfort working with Kubernetes.
- Experience building or operating complex distributed systems.
- Experience with multi-terabyte or multi-petabyte datasets.
- Comfort with large-scale data processing tools and moving between research code and production-quality systems.
Nice to have
- Experience with JAX, CUDA kernels or GPU workload optimization, or low-level Linux or kernel programming.
- Experience with large-scale video or multimodal datasets, training or evaluation infrastructure, distributed training, or deploying models into real-world systems, especially robotics.
Tools
Kubernetes, JAX, CUDA, Linux
AI focus
Train increasingly capable general models from internet-scale video, spanning multimodal representation learning and reinforcement learning, with eventual deployment to robots.
Model-written summary · the original posting is authoritative