torcrobotics
Software for autonomous heavy-duty trucking.
Company facts
- Founded
- 2005
- Employees
- 867
- Headquarters
- United States
- CEO
- Peter Vaughan Schmidt
- Type
- Private
- Funding
- Corporate Round · 2019
- Subsidiaries
- Algolux, Torc Europe GmbH
About
What began in 2005 as Torc Technologies, a venture by a group of Virginia Tech graduate students, has evolved into a leading name in autonomous vehicle technology. The mission from the outset was to continue their collegiate work on autonomous vehicle software, initially focusing on Tele-Operated Robotic Controls (TORC) and developing Level 4 autonomous technology. A significant early milestone came in 2007 when Torc, in partnership with Virginia Tech, secured third place in the DARPA Urban Challenge. This achievement involved their Ford Escape, 'Odin,' autonomously navigating 60 miles of urban and off-road environments. The company's journey continued with impactful projects, including a 2010 partnership with Virginia Tech for the National Federation of the Blind's Blind Driver Challenge, which earned them the National Instruments' 2010 Application of the Year. This project saw a blind driver independently operate Torc's modified Ford Escape on the Daytona Speedway in 2011. Torc's expertise also extended to defense, developing autonomous solutions like the Ground Unmanned Support Surrogate (GUSS) to assist military personnel. Over the years, Torc has been involved in various defense and heavy equipment applications, and participated in the DARPA Robotics Challenge. A pivotal moment arrived in March 2019 when Daimler AG, through Daimler Trucks North America, acquired a majority stake in Torc Robotics. This partnership shifted Torc's primary focus to commercializing autonomous trucks for long-haul applications in the United States. Torc is now an independent subsidiary of Daimler Truck AG. The company has since expanded its operations, opening an engineering office in Austin, Texas, and a Technology and Development Center in Stuttgart, Germany in 2022, as well as facilities in Albuquerque, New Mexico, and Montreal, Canada. Torc continues to advance its 'physical AI,' enabling self-driving trucks to perceive, understand, and act in the real world, with a commercial launch targeted for 2027.
ML Engineer, II - Perception
Build and scale Torc Sim, a simulation platform that replays and evaluates autonomy models against real and synthetic driving data, enabling teams to debug and improve ML models for self-driving trucks.
Senior, ML Engineer - Auto Tagging
Build ML pipelines to extract and tag rare driving scenarios from petabytes of truck sensor data, enabling safer autonomous trucking through structured scenario libraries.
ML Engineer, II - App Engine
Develops a C++ SDK/framework for composing and running autonomous-truck software on embedded Nvidia devices, focusing on performance, safety, and resource constraints.
Senior, ML Engineer - 3D Reconstruction
Build and deploy ML systems that reconstruct 3D lane lines and maps from noisy sensor data (cameras, LiDAR, radar, GPS/IMU) to power autonomous truck perception and HD mapping.
Senior, ML Engineer - VLM
Senior ML Engineer builds and curates large-scale vision-language datasets to train Torc’s autonomous truck-driving models, using cloud pipelines, open-vocabulary labeling, and distributed compute.
Senior, Machine Learning Engineer - 3D Perception
Senior Machine Learning Engineer designing and deploying 3D perception models for autonomous trucks, using PyTorch and multi-modal sensor data to improve environmental understanding.
Staff, ML Engineer - Scene Generation
Lead a team building neural rendering and generative models for autonomous-vehicle sensor simulation, translating research into production-grade camera, LiDAR, and radar pipelines.
Machine Learning Engineer, II - 3D Perception
Build and improve 3D perception models (LiDAR, cameras) for autonomous trucks using PyTorch and Python, integrating ML into Torc’s autonomy stack.
Senior, Machine Learning Engineer - Camera Model
Develop and deploy deep learning models for camera-based perception in autonomous trucks, focusing on object detection, segmentation, and depth estimation using Python and PyTorch.