About the role
You'll build the software that turns a capable robot into a product customers rely on, from building the Claude code like an agentic harness to the cloud services and data infrastructure down to the robot software itself.
Responsibilities:
- On-robot software and hardware integration: Build the ROS-based stack for perception, control, and task execution. Bring up and integrate robot arms, sensors, cameras, and teleop rigs.
- Physical agentic framework: Build agents that turn a job into a plan of steps, execute each one, verify it against real operating procedures, and escalate to a human when unsure. Build in tool use, memory, failure recovery, and human-in-the-loop handoffs.
- Cloud planners and fleet orchestration: Build services that break work orders into tasks, schedule them across robot fleets, and track progress and exceptions in real time.
- Vertical integrations and product: Connect our agents to each industry's existing systems, such as data center management, ticketing, maintenance workflows, and manufacturing execution systems. Build customer-facing dashboards, job setup, and audit trails.
- Data infrastructure and tooling: Build ingest, storage, indexing, and retrieval for large volumes of multimodal robot data. Build the tools our data operations team uses daily for capture, review, and annotation.
- Training and evaluation infrastructure: Partner with ML engineers on job orchestration, compute management, and reproducible pipelines.
Requirements:
- 5+ years of experience building and shipping production software, with a B. Tech, M. Tech, or PhD in computer science, electrical engineering, or a related field.
- A track record of owning a system end to end: you've designed it, shipped it, operated it, and been paged for it, and you can tell us what you'd do differently.
- Strong fundamentals in linear algebra, data structures, algorithms, operating systems, databases, and distributed systems design.
- Fluency in Python and at least one of C++, Rust, Go, or TypeScript, with real depth in whichever you'd call your main language.
- Experience with backend services, APIs, and cloud infrastructure (AWS or GCP, containers, databases) in a production setting.
- Strong software engineering habits: you write code others can build on and hold a high bar for code quality.
Preferred:
- Experience with ROS 2, real-time, or embedded software shipped to hardware in the field.
- Experience bringing up robot hardware: drivers, sensor and camera integration, and calibration.
- Experience building production systems with LLMs and agents: tool use, planning, and evaluation.
- Experience deploying ML models to edge compute (Jetson, ONNX, TensorRT, quantization).
- Experience with ML infrastructure: training orchestration (Kubernetes, Slurm, Ray), GPU clusters, experiment tracking, or model serving.
- Experience with large-scale multimodal data pipelines (object storage, Ray or Spark, Parquet, and dataset versioning).
- Experience with robot simulation at scale (Isaac Sim, MuJoCo).
- Experience in robotics, autonomous vehicles, or industrial automation.
- Experience mentoring engineers or leading technical efforts.