Senior / Staff Applied Scientist

Waabi Waabi · Robotics · Toronto, ON +3 · Remote · Autonomy & Algorithms

The role focuses on designing and building scalable production frameworks for evaluating autonomous driving systems, measuring the performance and reliability of the evaluation infrastructure, and developing analytical models to correlate simulation with real-world driving outcomes. It sits at the intersection of Evaluation, Systems & Safety, and Autonomy teams.

What you'd actually do

  1. Design scalable production frameworks for autonomy evaluation. You will apply statistically rigorous methodologies to challenges such as sampling evaluation sets to ensure comprehensive Operational Design Domain (ODD) coverage.
  2. Design and implement methodologies to systematically measure and monitor the performance, fidelity, and reliability of the evaluation infrastructure itself, ensuring our testing ecosystem remains a highly trusted source of truth.
  3. Directly prototype complex deep-dive analyses, write production-quality evaluation code, and build the sophisticated tools and dashboards that synthesize our progress for technical and executive leadership.
  4. Act as a key partner to Safety and Autonomy teams, ensuring that our automated reporting faithfully captures the technical intent and requirements of our system.
  5. Develop advanced analytical models to help correlate closed-loop simulation performance with real-world driving outcomes, quantifying the fidelity and predictive power of our evaluation metrics.

Skills

Required

  • Python
  • SQL
  • statistical methods
  • hypothesis testing
  • A/B testing on complex systems
  • sampling methodologies
  • statistical modeling
  • working with internal cross-functional partners/stakeholders

Nice to have

  • Autonomous Vehicle or Robotics industry experience
  • Systems Verification and Validation
  • large scale databases and analytics
  • workflow automation/orchestration frameworks

What the JD emphasized

  • rigor and reliability
  • rigorous frameworks
  • statistically rigorous methodologies
  • highly trusted source of truth
  • production-quality evaluation code
  • automated reporting faithfully captures the technical intent and requirements

Other signals

  • design scalable production frameworks for autonomy evaluation
  • design and implement methodologies to systematically measure and monitor the performance, fidelity, and reliability of the evaluation infrastructure
  • develop advanced analytical models to help correlate closed-loop simulation performance with real-world driving outcomes