[JOB] Researcher in Computer Vision / ML for human driver understanding and interactions
Guy Rosman
(Pls excuse cross-postings..)
Our team in TRI is seeking a research scientist for research at the intersection of machine learning, computer vision, and human factors. The role focuses on multimodal ML approaches for understanding, detecting, and developing intervention strategies for driver impairments, such as cognitive distraction and intoxication. The ideal candidate will contribute to fundamental research, publish in top-tier venues, and build machine learning models and prototypes that integrate human-in-the-loop data towards novel approaches for understanding and assisting drivers under diverse situations. The candidate will be part of a team of computational and cognitive researchers, devising and testing ML approaches on a variety of data sources and human-in-the-loop experiments.
This is an opportunity to work on cutting-edge research in human-robot interaction and intelligent vehicle systems in a collaborative and interdisciplinary team of experts in robotics, AI, and human factors. You will have access to state-of-the-art robotic platforms and simulation tools with the potential to contribute to academic publications and impactful real-world applications.
Apply at: https://jobs.lever.co/tri/89f860cc-13bd-4332-81f6-c2528d20a100
Or contact me for questions.
Thanks,
Guy Rosman
------
Responsibilities
Conduct original research on driver impairment detection and intervention (e.g. warning, coaching, actuation) using machine learning and computer vision.
Develop algorithms and models to analyze driver behavior, physiological signals, and other multimodal inputs.
Design, implement, and conduct human-in-the-loop behavioral studies, ensuring robustness and real-world applicability.
Publish findings in high-impact conferences and journals.
Collaborate with interdisciplinary teams, including human factors experts, cognitive scientists, and engineers.
Prototype and validate ML-based intervention strategies to enhance driver safety and performance.
Qualifications
PhD in Computer Vision, Machine Learning, Human-Centered AI, or a related field.
Research experience in human and machine vision, behavior analysis, or multimodal learning.
Strong publication record in top-tier conferences and journals (e.g., CVPR, NeurIPS, ICCV, ICLR).
Experience working with human-in-the-loop data: data collection, annotation strategies, and model training.
Proficiency in deep learning frameworks (e.g., PyTorch, Jax, Hugginface) and data analysis tools.
Ability to work both independently and as part of an interdisciplinary team.
Preferred Qualifications
Experience in developing real-time AI systems for human monitoring.
Familiarity with physiological and cognitive state estimation (e.g., eye tracking, EEG, heart rate variability).
Background in human factors, cognitive psychology, or related disciplines.
Experience deploying machine learning models in real-world environments.
Knowledge of software development best practices (version control, CI/CD, documentation).