Senior ML Validation Engineer

Job Description

The Senior ML Validation Research Engineer will lead applied machine learning research focused on improving verification and validation of ML components used in robotics and autonomous driving systems. This role centers on simulation-based evaluation, uncertainty modeling, scenario coverage automation, and transforming advanced ML research into working prototypes that enhance the efficiency, accuracy, and coverage of ML system validation. Key Responsibilities

Prototype research concepts into performant tools integrated into CI/CD and large-scale validation pipelines.

Advance ML research for open and and closed loop simulation validation.

Develop scenario generation, coverage-guided testing, and rare-event discovery tooling.

Create robust metrics, predictors, uncertainty and Out-of-Distribution detection methods for autonomy ML systems.

Evaluate deep learning modules across perception, prediction, and planning in realistic sensor and traffic simulation.

Improve behavioral coverage and hazard-aligned metrics used in release readiness decision making.

Collaborate with Simulation, Safety, Systems Engineering, and cross-functional partners.

Author technical documentation, white papers, and contribute to validation methodology standards.

Research Focus Areas

Scenario synthesis (diffusion models, generative models, counterfactuals)

Coverage-based and fuzzing-based evaluation for autonomy behavior

Uncertainty estimation, calibration, conformal prediction, OOD detection

Robustness testing and perturbation frameworks

Test suite prioritization, failure mining, and regression analysis

Required Qualifications

MS + 5 years, or PhD + 3 years in ML, Robotics, Computer Science, or work related experience

Experience with simulation-driven ML evaluation for robotics/autonomy

Strong proficiency in Python, PyTorch/JAX/TensorFlow

Demonstrated ability to translate complex ML research ideas into functional prototypes

Experience integrating ML evaluation into CI/CD pipelines

Proven research impact through published work, internal tools, or patents

Strong communication skills and ability to collaborate cross-functionally

Preferred Qualifications

Experience with autonomy stacks (perception/prediction/planning).

Familiarity with CARLA, SVL, DriveSim, Applied Intuition, or equivalent simulation platforms.

Knowledge of Bayesian ML, causal inference, and sequential testing.

Experience with digital twin systems and sensor simulation.

Understanding of automotive safety standards (ISO 26262, UL 4600, SOTIF).

Experience building validation dashboards and scorecards connected to release criteria.

Success Criteria

Faster detection of ML regressions with improved test efficiency

Improved