Dyna Robotics
Software Engineer, Data Infra
About this role
Build the data infrastructure powering Dyna's AI-driven robots. You'll architect systems that handle multimodal sensor streams, enable human-in-the-loop workflows, and transform raw robot data into production-ready training signals for advanced generalist robotic models.
What you'll do
- Design interactive systems to unify robot logs, video, and 3D sensor data with seamless annotation and analysis interfaces
- Develop algorithms to extract structured signals like trajectories and 3D poses from raw multimodal captures
- Build evaluation tools to benchmark model-driven motion against human-captured ground truth
- Implement scalable distributed data pipelines for ingestion and transformation of terabytes of data
- Create visualization and debugging tools for model behaviors and sensor data interpretation
- Collaborate across ML, robotics, and product teams to define data infrastructure roadmap
What they're looking for
- Python (NumPy, Pandas)
- Geometric algorithms and 3D spatial reasoning
- Relational and NoSQL databases (PostgreSQL, Redis)
- Cloud infrastructure (GCP/AWS, Kubernetes)
- React/TypeScript for internal tools
- Distributed data pipeline systems
- Multimodal data handling (video, LiDAR, time-series)
- Debugging and optimization of data systems
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Dyna Robotics
Dyna Robotics builds AI-driven robotics platforms that combine embodied AI foundation models with spatial intelligence to enable robots to navigate and operate autonomously across customer sites. The company is hiring software engineers, infrastructure specialists, QA testers, and field deployment engineers to scale its robot deployment systems, training infrastructure, state estimation capabilities, and customer operations.
View all jobs at Dyna RoboticsLikely interview questions
- Walk us through your experience building data pipelines at scale. How have you handled multimodal data (video, sensor streams, time-series) in production?
- Describe a time you debugged a complex data quality issue in a production system. How did you isolate the root cause and what tools did you use?