Senior Machine Learning Engineer - Perception

The Senior Machine Learning Engineer, Perception, will lead perception and grasping machine learning efforts end-to-end, from defining the problem through dataset creation, training, and evaluation. The role focuses on the modeling loop, while platform and robotics teams own runtime integration and edge deployment.

Responsibilities

• Own ML problem formulation: translate ambiguous picking performance goals into well-posed ML tasks, including selecting targets, label and reward definitions, the optimization loss, and the criteria for what “better” means.
• Build the data engine: manage dataset strategy across the full lifecycle, including large-scale synthetic scene generation, distributed cloud dataset jobs, scene diversity filtering, and self-supervised label generation.
• Identify data bottlenecks: determine when performance is constrained by data rather than the model, and implement fixes to improve dataset quality and coverage.
• Train and iterate on large vision models: fine-tune large backbones using multiple task-specific heads, run disciplined experiments, and decide both the next experiment to run and when to stop.
• Own evaluation and verification: define offline metrics that correlate with on-robot outcomes, and use online metrics to improve performance on edge cases.
• Drive technical excellence: perform rigorous code reviews, mentor engineers earlier in their careers, and contribute to the perception and grasping roadmap.
• Hand off cleanly to production: deliver deployable models with practical latency budgets, exportable graphs, and reproducible artifacts, in partnership with engineers responsible for runtime integration and inference optimization.

Required Qualifications

• 5+ years of experience in machine learning and computer vision, including carrying models from problem statement to measured outcomes.
• Track record of framing problems with clear, measurable definitions.
• Hands-on experience building datasets at scale: generation or collection, curation, automated quality assurance, coverage and bias analysis, and active learning to systematically address edge cases.
• Rigorous evaluation habits: use baselines, ablations, metric design, and strong judgment on whether results are real.
• Expert-level Python and PyTorch skills.
• Comfort with cloud training infrastructure including GCP or AWS, plus Docker and experiment tracking tools.
• Professional experience with one or more of the following: multi-task model development and training; multi-modal model development and training; multi-view model development and training; self-supervised, outcome-based, or reward-driven learning.

Technologies

• Python
• PyTorch
• GCP
• AWS
• Docker

Location and Compensation

• Location: Charlestown, MA (onsite)
• Salary: USD 170,100 - 227,400 per year

Benefits

• Health, dental, and vision insurance
• Unlimited vacation
• Paid federal and state holidays
• 401(k) contributions of 5% of salary
• Company-covered travel expenses
• Other perks to support day-to-day work comfort and productivity

AI Tools Notice

• The company may use AI tools to support parts of the hiring process, such as reviewing applications, analyzing resumes, or assessing responses and identifying potential inconsistencies or verification signals based on available information.
• These tools assist the recruitment team and do not replace human judgment.
• Final hiring decisions are made by humans.
• If you want more information about how data is processed, contact the company.