Robotics Controls Engineer

TerranovaBerkeley, CA
3h

About The Position

Backed by leading climate and American dynamism investors, Terranova builds intelligent robotic systems to terraform the Earth itself - lifting land, restoring wetlands, and protecting critical infrastructure from floods and sea-level rise. Our mission is to preserve the built environment, create new habitats, and usher in an era of abundance. Our work supports climate resilience, disaster recovery, and defense across the United States and beyond. We’re assembling a world-class team that wants to work on something real, physical, and civilization-scale. If you want your work to reshape the world (literally), this is the place to do it. What to to Expect We’re seeking a robot controls and machine learning specialist to develop algorithms that give our robotic systems adaptive, intelligent behavior. You’ll design and tune controllers, build dynamic models, and integrate them with perception and sensor data. This is a chance to bridge classical control with modern machine learning to shape how machines move through the Earth.

Requirements

  • Bachelor’s degree or higher in Computer Science or related field
  • U.S. permanent residency required

Nice To Haves

  • Python/C++ with PyTorch/JAX
  • MPC/OSQP/CasADi
  • EKF/UKF/factor graphs
  • System ID, RL (PPO/SAC) with safety shields
  • ROS2, ONNX/TensorRT
  • Latency and jitter profiling
  • PCB and embedded systems design skills are a huge bonus
  • Have worked on large complex robotic systems - various sensors, multiple motors/robots, reinforcement learning based on robot data
  • Maker background - love building stuff, work is fun to you
  • Coming from a startup environment - high ownership, fast paced environment

Responsibilities

  • Develop and tune controllers (PID, MPC, optimal/robust control) for dynamic, nonlinear systems.
  • Build and validate physical models for simulation, hardware-in-the-loop testing, and autonomy.
  • Train and deploy ML models for perception, planning, or adaptive control (supervised or RL).
  • Integrate algorithms with firmware and cloud teams, ensuring real-time safety and stability.
  • Profile, optimize, and verify performance under latency, jitter, and compute constraints
  • Comfortable with the pace and intensity of early-stage startup life, including long days, 6-day workweeks, and extended field hours
© 2024 Teal Labs, Inc
Privacy PolicyTerms of Service