Senior Autonomy Engineer

Brain Corp•San Diego, CA

1d•Onsite

About The Position

As a Senior Autonomy Engineer on our R&D team, you'll help define the next generation of software that lets robots perceive, learn, and act in unstructured indoor environments. You'll work across the modern autonomy stack — from learned perception and prediction to mapping and motion planning — and ship capabilities that generalize across our fleet. We're looking for someone equally comfortable reading a fresh arXiv paper, writing production C++/Python, and debugging behavior on a real robot in the lab. You'll help set technical direction, raise the bar on engineering quality, and mentor others on the team.

Requirements

Master’s Degree. or Ph.D. in Computer Science, Robotics, Electrical Engineering, or a related field — or equivalent demonstrated experience.
5+ years of relevant industry or research experience building autonomy, perception, or ML systems
Strong fluency in Python and C++ in a Linux environment.
Demonstrated track record of taking research ideas and papers into deployed implementations.
Depth in one or more of: machine learning (supervised, self-supervised, imitation, RL), SLAM and state estimation, motion planning, or 3D perception.
Hands-on experience with PyTorch (and/or JAX), modern training pipelines, and contemporary architectures (transformers, diffusion models, vision-language-action models).
Experience designing robotic systems with ROS 2 (or comparable middleware) and contemporary simulation tools such as Isaac Sim, MuJoCo, or Gazebo.
Comfort with the full ML lifecycle: data curation and labeling strategy, large-scale training, offline and online evaluation, and continuous deployment to production hardware.
Solid systems and software architecture instincts; pragmatism about when a learned approach beats a classical one and vice versa.
Familiarity with modern engineering practices: CI/CD, code review, observability, and iterative delivery (i.e. Agile, Scrum).

Nice To Haves

Contributions to open-source robotics or ML projects, publications at top venues (CoRL, RSS, ICRA, NeurIPS, CVPR, ICML), or experience with on-device acceleration (TensorRT, ONNX, custom CUDA kernels).

Responsibilities

Design, train, and deploy machine learning systems for perception, SLAM, prediction, and motion planning that enable safe navigation around people and obstacles.
Build and improve learned components — including transformer-based perception, vision-language models for scene understanding, diffusion or imitation-learning policies, and neural/Gaussian-splatting approaches to mapping — and integrate them with classical estimation and planning where it makes sense.
Translate state-of-the-art research (papers, open-source releases, conference talks) into production-quality implementations on the robot.
Develop data engines and evaluation infrastructure that turn fleet logs into training data, regression tests, and shipped improvements.
Own features end-to-end: from prototype, through sim and on-robot validation, to fleet rollout, with measurement of real-world impact.
Improve runtime performance of perception, mapping, and planning on embedded GPU/accelerator hardware (quantization, distillation, kernel work where warranted).
Contribute to internal frameworks, simulation tooling, and developer experience that compound team velocity.
Provide guidance and mentorship to engineers across robotics, ML, and the software systems that support them.