Staff, ML Engineer - Learned Behaviors / RL Planning

📋 Description

• • As a Staff Machine Learning Engineer specializing in Learned Behaviors and Reinforcement Learning (RL) for Planning at Torc Robotics, you will be at the forefront of developing and deploying sophisticated AI models that enable autonomous trucks to make safe, efficient, and human-like driving decisions. This pivotal role focuses on the core innovation and maturity of learned behavior models, rather than their downstream integration into specific features. You will be instrumental in shaping the future of autonomous freight transportation by leading the charge in creating intelligent systems that can navigate complex real-world scenarios.
• • Your primary responsibility will be to architect, develop, and rigorously validate learned behavior models. This includes, but is not limited to, techniques such as driver mimicry, multi-agent interaction modeling, behavior cloning, and reinforcement learning. These models are crucial for enabling autonomous trucks to operate safely and efficiently on highways and in freight operations, requiring a deep understanding of how vehicles interact with their environment and other road users.
• • A significant aspect of this role involves defining and implementing robust data strategies. You will be responsible for the collection, meticulous labeling, and curation of vast datasets. These datasets will encompass a wide range of information, including in-vehicle driving data, simulation logs, and fleet operational data, all of which are essential for training and evaluating the performance of your models.
• • You will design and build scalable model training pipelines, along with the necessary infrastructure and tooling. This will empower the team to iterate rapidly on behavior models, moving them efficiently from initial prototypes to production-ready systems. The ability to develop and manage these pipelines is key to accelerating the development cycle and ensuring timely deployment of advanced AI capabilities.
• • A critical function of this role is to design, implement, and track key performance metrics for the learned behavior models. You will be tasked with analyzing model behavior, diagnosing failure modes, and driving continuous improvement. This iterative process will involve extensive testing in both simulation environments and on actual vehicles to ensure the models perform optimally under all conditions.
• • Collaboration will be paramount. You will work closely with simulation, scenario generation, and validation teams to seamlessly integrate the learned behavior models into comprehensive verification and validation frameworks. This ensures that the models are thoroughly tested and meet the required safety and performance standards across all Operational Design Domains (ODDs).
• • As a Staff Engineer, you will also take on a leadership and mentorship role. You will guide and mentor mid-level and senior ML engineers, setting the technical direction for the team. Furthermore, you will champion and drive best practices in learned behavior engineering, fostering a culture of excellence and innovation within the team.
• • This role is deeply technical and requires a hands-on approach to problem-solving. You will be expected to contribute directly to the development and implementation of cutting-edge ML techniques, pushing the boundaries of what's possible in autonomous driving. Your work will have a direct impact on the safety, efficiency, and scalability of Torc's autonomous trucking solutions.
• • The opportunity at Torc offers a unique chance to work on a challenging and impactful problem in the rapidly evolving field of autonomous vehicles. You will be part of a pioneering company, a part of the Daimler family, and contribute to transforming the future of freight transportation. Your expertise will be vital in developing the learned behaviors that will define the next generation of self-driving trucks, making them safer, more reliable, and more efficient than ever before.