Staff, ML Engineer - E2E

About the Role

At Torc, we have always believed that autonomous vehicle technology will transform how we travel, move freight, and do business. A leader in autonomous driving since 2007, Torc has spent over a decade commercializing our solutions with experienced partners. Now a part of the Daimler family, we are focused solely on developing software for automated trucks to transform how the world moves freight. Join us and catapult your career with the company that helped pioneer autonomous technology, and the first AV software company with the vision to partner directly with a truck manufacturer. As a Staff Machine Learning Engineer focused on End-to-End (E2E) Model Development, you will lead the design and deployment of learning-based architectures that connect perception inputs to driving decisions — advancing the frontier of closed-loop autonomous driving performance. You’ll architect and push forward Torc’s End – to – End approaches through unified, differentiable pipelines, leveraging massive real-world and simulated datasets to continuously improve system intelligence. This is a high-impact technical leadership role focused on core model research and large-scale ML development, not feature-layer logic or rule-based planning.

Responsibilities

• Lead E2E model design and development — define architectures that directly map multi-modal sensor inputs (camera, LiDAR, radar, HD maps) to mid- or high-level driving actions or cost functions.
• Drive large-scale training and evaluation for E2E learning, integrating data from perception, behavior prediction, and control systems.
• Develop and refine learning objectives that align with real-world driving metrics: safety, comfort, compliance, and efficiency.
• Architect scalable pipelines for multi-task, multi-modal learning, leveraging both real-world and synthetic data.
• Prototype and evaluate new paradigms such as differentiable planning, imitation learning, reinforcement learning, and world models for AV behavior.
• Collaborate cross-functionally with Perception, Prediction, and Motion Planning teams to align interfaces and ensure consistency between learned and modular components.
• Establish robust evaluation frameworks for E2E performance, including closed-loop simulation and on-road validation.
• Mentor engineers and scientists in large-scale experimentation, model interpretability, and data-driven debugging.
• Stay at the frontier of ML research, exploring advancements in foundation models, sequence modeling, self-supervision, and generative world representations.

Requirements

• 10+ years of experience developing deep learning systems for perception, planning, or control.
• M.S. or Ph.D. in Computer Science, Robotics, Electrical Engineering, or related field (or equivalent practical experience).
• Deep expertise in multi-modal ML, sequence modeling, or policy learning (e.g., Transformers, diffusion models, imitation learning).
• Proven track record in large-scale model training and optimization for real-world tasks.
• Strong proficiency in Python, PyTorch, or TensorFlow, and experience with distributed ML frameworks.
• Solid understanding of sensor fusion, spatiotemporal modeling, and vehicle dynamics.
• Demonstrated leadership in driving technical roadmaps, mentoring teams, and delivering production-quality ML solutions.

Qualifications

• Experience developing E2E or mid-to-end models for autonomous driving, ADAS, or robotics.
• Familiarity with differentiable cost maps, latent space planning, or behavior cloning / reinforcement learning in driving domains.
• Hands-on experience with simulation-in-the-loop training and evaluation.
• Understanding of safety validation and interpretability for learned driving systems.
• Publications or open-source contributions in top-tier ML or robotics venues (CVPR, NeurIPS, ICLR, ICRA, CoRL).
• Experience with foundation models or large-scale multimodal pretraining for perception and planning
• Experience using Ray