Torc Robotics
Senior, ML Engineer - Auto Tagging
$177K–$213K
Ann Arbor, MI, Remote - US +1 · Remote · Senior · Full-time · Posted May 20, 2026
- Location
- Ann Arbor, MI, Remote - US
Ann Arbor, Michigan, United States - Experience
- 6+ years with a Bachelor’s degree, 3+ years with a Master’s degree, or 1+ years with a PhD.
- Education
- Bachelor’s, Master’s, or PhD in Computer Science, Robotics, Electrical Engineering, or a related technical field.
About the job
Builds systems that mine large-scale autonomous-truck data for driving scenarios, tag and structure those events, and make them available for downstream machine learning, testing, and validation.
What you'll do
- Architect distributed pipelines that extract and catalog safety-critical and long-tail events from multi-sensor vehicle logs.
- Develop heuristic and ML-assisted algorithms to classify and describe scenarios, and integrate tagged events into the observations database.
- Coordinate with downstream teams and mentor engineers through design reviews, coding standards, and technical guidance.
What they ask for
- Bachelor’s degree in a related technical field plus 6+ years of experience, or Master’s plus 3+ years, or PhD plus 1+ years.
- Strong Python and SQL skills and experience processing massive time-series or unstructured datasets.
- Hands-on machine learning and dataset curation experience; Databricks or similar large-scale analytics platforms.
- Expertise in distributed compute frameworks and cloud platforms; experience with scenario-description standards like Pegasus layers.
Nice to have
- Familiarity with foundational models, auto-labeling pipelines, or zero-shot classification.
- Experience with vLLM, SGLang, robotics data formats, columnar storage, or semantic retrieval systems.
Pay
$177,300 — $212,800 USD (base salary)
Tools
Python, SQL, Databricks, Ray, Spark, Beam, AWS, GCP
AI focus
Develops ML-assisted scenario tagging and dataset curation; may explore Vision-Language Models, semantic vector search, auto-labeling, and semantic inference.
Model-written summary · the original posting is authoritative