1 day ago
Responsibilities
- Build and evaluate proof-of-concept autonomy skills and establish measurable performance baselines.
- Define architecture and interfaces across data, simulation, training, policy inference, controls, and evaluation.
- Develop, train, evaluate, and deploy reinforcement-learning, behavior-cloning, and imitation-learning policies.
- Lead cross-team execution, resolve technical blockers, mentor engineers, and implement critical system components.
- Build data collection, synchronization, curation, quality-checking, versioning, and reproducible training tools.
- Transfer policies from simulation to physical machines while addressing sensor noise, latency, action limits, inference performance, and changing dynamics.
- Integrate learned policies with perception, state estimation, controls, runtime supervision, fallbacks, and operating-mode transitions.
- Establish evaluation, regression testing, failure diagnosis, and controlled hardware validation for reliable field deployment.
Requirements
- Staff-level or above technical ownership of substantial robotics or robot-learning efforts, including technical direction, architecture decisions, cross-team coordination, and delivery of working systems.
- Demonstrated personal implementation, training, deployment, and evaluation of learned policies on physical robots or vehicles, with depth in reinforcement learning, imitation learning, or behavior cloning.
- Recent hands-on contributions to production or deployment-quality software, learning experiments, and hardware debugging.
- Experience mentoring engineers and improving technical decisions through design and code reviews.
- Strong understanding of machine learning, optimization, sequential decision-making, generalization, and distribution shift.
- Proficiency in Python and a modern learning framework such as PyTorch or JAX.
- Experience building reproducible data, training, and evaluation workflows and interpreting experimental results.
- Working knowledge of robotics, including observations, actions, dynamics, feedback, and physical constraints.
- Ability to write maintainable software, diagnose failures across system boundaries, and communicate technical tradeoffs clearly.
- Degree in Robotics, Computer Science, Electrical or Mechanical Engineering, or a related field, or equivalent practical experience.
- Preferred experience includes sustained field or customer deployment, demonstration collection, teleoperation, dataset design, offline reinforcement learning, policy fine-tuning, simulation-to-reality transfer, system identification, domain randomization, visuomotor or diffusion policies, hierarchical or multitask skill learning, C++, real-time robot software integration, and robotics applications involving variable loads or contact conditions.
Categories
ML EngineeringRobotics
About AIM
AIM builds autonomy for heavy equipment, turning bulldozers, excavators, and loaders into AI-driven fleets for mining, construction, infrastructure, and defense operators. The company sells an autonomy platform combining software, machine controls, and ruggedized hardware, deployed on active jobsites to run large-scale earthmoving. Founded in 2021 and headquartered in Redmond, Washington, AIM is privately held and backed by investors including General Catalyst and Khosla Ventures.