Pantograph PBC
Infrastructure & Systems Lead
San Francisco · On-site · Full-time · Posted Jul 14, 2026
- Education
- No specific credential required; the posting prioritizes what candidates have built.
About the job
Architect and operate the end-to-end infrastructure pipeline connecting a large robot fleet to inference and training clusters, under real-time latency constraints.
What you'll do
- Own architecture decisions across networking, storage, databases, embedded software, deployment systems, and GPU optimization.
- Build infrastructure for robot fleets communicating with inference clusters and streaming video to training clusters.
- Support frequent deployment of new model weights within strict real-time latency budgets.
What they ask for
- Designed and operated large-scale distributed systems from scratch.
- Managed fleets of hundreds or thousands of computers.
- Deep GPU performance optimization experience, including writing custom CUDA kernels.
- Experience in real-time embedded systems or robotics infrastructure, petabyte-scale storage and database systems, or high-performance networking and video encoding pipelines.
Nice to have
- Rust and low-level performance optimization experience.
- Experience taking infrastructure from prototype to production in a small team.
Tools
CUDA, Rust
AI focus
Infrastructure for robot systems that use models trained on internet-scale video, games, and real-world video; includes inference and frequent model-weight deployment.
Model-written summary · the original posting is authoritative