Perception Algorithm Engineer

What you'll do

• Design and implement robust, real-time multi-object tracking systems for computer vision problems.
• Utilize multimodal estimates from vision and audio to ensure high-fidelity estimation in complex environments.
• Develop rigorous evaluation frameworks and curate datasets to benchmark model performance.
• Define metrics to analyze edge cases and continuously improve perception pipelines.
• Integrate perception systems into a larger software stack while meeting real-world performance constraints.
• Write performant, production-quality code using C++, Swift, or Python.
• Implement both modern deep learning models and traditional state-estimation pipelines.

What we're looking for

• A PhD in Computer Science, Robotics, or a related field with 3 years of industry experience is required.
• An MS degree in a related field with at least 2 years of industry experience is required.
• Proficiency in systems programming using C++ or Swift to write performant, production-quality code is required.
• Fluency in Python and modern machine learning frameworks like PyTorch or JAX is required.
• A solid foundation in machine learning, traditional perception, and state-estimation pipelines is required.
• Experience building or deploying on-device computer vision models or multi-object tracking systems is preferred.
• Expertise in advanced ML architectures such as VLMs, VLAs, foundation models, or self-supervision is preferred.
• Knowledge of robotics software stacks, including SLAM, kinematics, and reinforcement learning, is preferred.