Founding Computer Vision & Perception Engineer
Location: San Francisco
1. About the Company
TerraWise is building autonomous ground robotics and AI to power the next generation of utility-scale solar operations. As solar deployments scale rapidly, maintaining vast, remote infrastructure has become a critical challenge. Today’s O&M teams rely on labor-intensive inspections and fragmented data to keep sites running. TerraWise changes that by deploying rugged autonomous rovers paired with a real-time intelligence platform that continuously monitors assets, detects issues early, and delivers actionable insights. Our mission is to augment field teams; improving uptime, safety, and efficiency while reducing operational costs.
We’re a venture-backed, Seed-stage deep-tech startup at the intersection of robotics, AI, and clean energy, building systems designed to operate in harsh real-world environments, solving real problems for a positive impact. If you want to work on technology that directly accelerates the world’s transition to renewable energy, we’d love to talk.
2. About the Role
What you’ll do
- Build and own the end-to-end perception pipeline: sensor ingestion, calibration, fusion, and inference.
- Design and train models for object detection, semantic/instance segmentation, multi-object tracking, 3D reconstruction tasks
- Own full model lifecycles: dataset curation, labeling strategy, training, evaluation, and continuous improvement
- Close the loop from robot sensors → CV inference → actionable insights on TerraWise's customer platform.
Requirements
- M.S. or Ph.D. in CS/EE/Robotics focused on Computer Vision and AI, or equivalent industry experience
- 2–4+ yrs building and deploying CV/perception systems in production or research
- Strong programming background with Python, Pytorch, Tensorflow, and other ML tools
- Hands-on experience with multimodal sensing (thermal, RGB, depth, point cloud, etc.) including cross-modal fusion and calibration
- Proven ability to build reliable models with limited labeled data with few-shot, semi-supervised, or self-supervised approaches
- Experience working with foundation models, VLMs
- Experience with Docker, AWS, SSH, CI/CD pipelines
- Ability to read, implement, and adapt SOTA methods from recent CV literature; stays current with the latest research and publications.
Bonus skills
- Robotics: ROS, C++
- Simulation: Gazebo, MuJoCo, IsaacSim
3. Benefits and Pay
- $125,000-$175,000 base salary
- Benefits: Health insurance, Gym membership, office lunches, etc.
- 0.25-1% equity on a standard 4-year vesting schedule with a 1-year cliff (stock options value $100K+)
- Start date: April 2026
