Job Description
About Gravis Robotics
Gravis Robotics is a startup turning heavy construction machines into intelligent and autonomous robots. Our unique combination of learning-based automation and augmented remote control enables a single operator to safely manage a fleet of earthmoving machines in a gamified environment. With over a decade of academic experience at the cutting edge of large-scale robotics, our team is rapidly translating this expertise into real-world deployments with industry leaders in a trillion-dollar market.
About the Job
At Gravis, the intelligence behind our machines is only as good as the systems that develop, train, and operate it. The Gravis Rack fuses data from LiDAR, cameras, GNSS, and hydraulics into a learning-based control system that adapts in real time to changing ground conditions. As our fleet grows and our models become more sophisticated, we need world-class infrastructure to support the full ML lifecycle: from raw sensor data ingestion on the edge to continuous model training, evaluation, and deployment at scale.
As our Data & ML Ops Engineer, you will be driving the requirements gathering, development, rollout and operation of the related infrastructure. The systems you build and operate power every ML experiment, training run, and production deployment at Gravis. You will work at the intersection of our Platform, Autonomy, and Perception teams to enable high velocity & quality ML development & deployment.
Responsibilities
-
Design and operate end-to-end ML pipelines covering data ingestion, preprocessing, versioning, training, evaluation, and deployment ranging from edge devices in the field to cloud training infrastructure
-
Build and maintain a scalable data platform for large-scale multimodal robotics datasets (LiDAR point clouds, camera imagery, GNSS/IMU, and other machine data)
-
Own CI/CD pipelines for ML workflows, including automated model training, regression testing, and staged deployment to production fleets
-
Manage experiment tracking, model registry, and artifact versioning to ensure full reproducibility across research and production
-
Collaborate closely with Autonomy and Perception engineers to understand requirements and translate them into reliable, scalable training environments
-
Evaluate and integrate best-in-class MLOps tooling on cloud and on-prem compute platforms
Required Qualifications
-
Bachelor's or Master's degree in Computer Science, Data Engineering, Electrical Engineering, or a related field
-
3+ years of hands-on experience in ML Ops, data engineering, or ML infrastructure roles
-
Strong Python skills and solid experience with ML frameworks such as PyTorch or TensorFlow
-
Proven experience building and managing CI/CD pipelines for ML workloads (e.g. GitHub Actions or GitLab CI)
-
Hands-on experience with containerization (Docker).
-
Experience with cloud platforms (AWS, GCP, or Azure).
-
Experience with data versioning, experiment tracking and workload orchestration tools (e.g. MLflow, W&B, clear.ml, DVC).
Bonus Qualifications
Don't meet every requirement? We still want to hear from you. These are nice-to-haves, not dealbreakers:
-
Experience with GPU accelerated simulation environments (e.g. IsaacSim/IsaacLab, CARLA, MuJoCo)
-
Experience working with robotics data (point clouds, camera streams, timeseries data).
-
Hands-on experience with infrastructure as code
-
Experience with Robotics & DevOps related tooling (Foxglove, Prometheus, Grafana)
-
Experience scaling ML infrastructure