2 months ago
Responsibilities
- Develop and iterate data miners that locate potential RotR violations while following SSO requirements.
- Develop and iterate VLM/LLM workflows and pipelines for classifying RotR events.
- Extend triagers from simulation data to fleet and service data.
- Curate golden datasets and evaluate triager precision and recall against human triage.
- Analyze error cases and translate findings into pipeline improvements.
Requirements
- Strong Python, PySpark, and SQL skills.
- Coursework in machine learning, computer vision, or NLP, with understanding of supervised learning, model evaluation, dataset bias, and label noise.
- Hands-on experience building production LLM/VLM applications, including agentic workflows, RAG, tool calling, evaluation, or fine-tuning.
- Experience evaluating multimodal or vision-language models.
- Currently enrolled in a relevant B.S. or M.S. program.
- Experience with prompt optimization, model fine-tuning, or human-in-the-loop ML systems.
- Familiarity with Git, unit testing, debugging, and code reviews.
- Strong Scala skills are a bonus.
- Experience with Spark optimization, distributed data-processing systems, autonomous vehicles, robotics, mapping, or transportation-related datasets is a bonus.
Benefits
- Minimum three-month assignment
- Minimum commitment of 40 hours per week
- On-site work at one of Zoox's office locations
Tech Stack
Categories
About Zoox
Zoox builds a fully autonomous, purpose-built electric vehicle and the end-to-end platform to operate it as an urban robotaxi service. The company designs the vehicle, self-driving software, and fleet operations in-house to offer mobility-as-a-service rather than selling cars. Founded in 2014 and headquartered in Foster City, California, Zoox is a subsidiary of Amazon.