About the Role
The position involves building and improving AI-driven agents that operate in simulated physical environments, enabling robots to learn complex behaviors and interact effectively with their surroundings.
Responsibilities
- Design and implement simulation environments for training AI agents
- Develop perception and control systems for embodied agents
- Optimize agent performance using reinforcement learning techniques
- Collaborate with robotics engineers to align simulation with real-world constraints
- Create realistic physics-based models for virtual testing
- Integrate sensor data into simulation pipelines
- Debug and refine agent behavior across diverse scenarios
- Evaluate generalization of trained policies from simulation to physical systems
- Improve simulation fidelity to reduce domain gap
- Support deployment of AI models on robotic platforms
- Contribute to software architecture for scalable simulation frameworks
- Develop tools for visualizing agent behavior and training progress
- Ensure simulation environments reflect real-world variability
- Work on multi-agent interaction scenarios
- Maintain and extend existing simulation codebase
- Implement safety protocols within virtual training environments
- Collaborate on defining success metrics for embodied tasks
- Assist in benchmarking simulation-to-reality transfer
- Refactor systems for improved computational efficiency
- Document simulation design decisions and implementation details
- Support automated testing of AI behaviors
- Iterate on environment design based on experimental feedback
- Work closely with machine learning researchers
- Ensure reproducibility of simulation experiments
- Contribute to version control and experiment tracking systems
Nice to Have
- Master’s or PhD in robotics, AI, or related field
- Published research in embodied AI, simulation, or robotics
- Contributions to open-source robotics or simulation projects
- Experience with ROS or similar robotic frameworks
- Work involving large-scale simulation training
- Background in behavioral cloning or imitation learning
- Familiarity with differentiable simulation
- Experience with real robot deployment challenges
- Knowledge of safety-critical system design
- Prior role in a robotics startup or research lab
Compensation
Competitive salary and equity package
Work Arrangement
Hybrid work model with flexible remote options
Team
Small, interdisciplinary team focused on robotics and AI development
Why This Role Matters
- This position plays a central role in bridging the gap between virtual training and physical robot performance, enabling faster and safer development cycles.
- Engineers in this role directly influence how robots learn to interact with complex, unstructured environments.
Technology Stack
- The team uses Python, PyTorch, MuJoCo, Unity, ROS, Docker, and custom C++ modules for high-performance simulation.
- Internal tools are built around scalable cloud infrastructure for distributed training.
Available for qualified candidates
