New: Building a Preschool for Robots
Pantograph is building generally intelligent robots that learn by doing: exploring, failing, and continuously improving. They teach themselves through self-supervised reinforcement learning: every trajectory they collect, successful or not, is something they learn from.
We own our entire stack, from hardware to firmware to training infrastructure and learning algorithms. Our hardware is designed from the ground up to be inexpensive enough for large-scale data gathering, and durable enough to survive the mistakes of the models.
If you're exceptional, and want to work on a small team at the intersection of hardware, learning, and scale, we're hiring:
Open Positions
- Research Scientist
- Research Engineer
- Mechanical Design Engineer
- Infrastructure & Systems Lead
- Assembly Technician
If you don't see a role that fits but our work resonates, we still want to hear from you. Reach out at hiring@pantograph.com with a CV, a description of what you excel at, or something that shows what you can do — a project, a paper, something you built.
About
Pantograph was founded by Alex Gajewski and Kelsey Pool. Alex previously cofounded SF Compute and Exa AI, helped organize the first batch of AI Grant, and worked with Ken Stanley and Joel Lehman on evolutionary RL. Kelsey led the design for a low-cost humanoid that became a KScale product, and before that, she designed robotic systems for CAR-T cell therapies at Multiply Labs that deployed at UCSF and Stanford.
Pantograph is a Public Benefit Corporation based in San Francisco.