Staff Machine Learning Engineer – Machine Learning, Road and Lane Detection

Job not on LinkedIn

Yesterday

Apply Now

Receive Emails with Similar Jobs

Torc Robotics

WebsiteLinkedInAll Job Openings

Transport • Hardware • Artificial Intelligence

Torc Robotics is an innovative company focused on commercializing self-driving trucks for long-haul freight transportation. As an independent subsidiary of Daimler Truck, the company is developing autonomous technology, primarily focusing on the Freightliner Cascadia. Torc is committed to safe transportation, continuously improving its solutions through rigorous testing and integration of industry-leading sensors. It collaborates with fleet management companies to deploy real-world autonomous solutions, aiming to lead the industry in autonomous trucking.

501 - 1000 employees

Founded 2007

🚗 Transport

🔧 Hardware

🤖 Artificial Intelligence

📋 Description

• Own the model roadmap for road and lane detection within the perception machine learning organization, from concept to production-grade model maturity. • Research, design, and train advanced neural architectures (e.g., multi-camera BEV transformers, LiDAR-vision fusion models, topological lane-graph networks) to detect, segment, and model roadway structures and lane connectivity. • Lead the data strategy for this domain: define data curation, labeling policies, and active learning pipelines to capture long-tail scenarios (e.g., occlusions, complex merges, construction zones). • Develop robust metrics and evaluation frameworks for lane and road geometry accuracy, temporal consistency, and cross-domain generalization. • Advance foundational capabilities such as self-supervised pretraining, synthetic-to-real adaptation, and temporal modeling for road and lane understanding. • Run large-scale experiments: design, execute, and analyze distributed training workflows and ablation studies to identify scalable improvements. • Collaborate with other perception teams to ensure model and interface consistency. • Mentor engineers and scientists, defining best practices for training, model evaluation, and code quality. • Stay at the cutting edge of research by evaluating and adapting emerging techniques (e.g., large BEV-based models, vectorized map prediction, lane-graph transformers) to production-grade perception.

🎯 Requirements

• 10+ years of experience developing deep learning models for perception or computer vision at scale. • Master’s or PhD in Computer Science, Electrical Engineering, Robotics, or a related field — or equivalent experience. • Deep expertise in semantic and instance segmentation, BEV modeling, or scene topology estimation. • Strong understanding of lane and road geometry modeling, camera calibration, and sensor projection. • Proficiency in Python and modern ML frameworks (e.g., PyTorch, Lightning). • Experience with distributed training pipelines, experiment management, and handling large-scale datasets. • Proven leadership in shaping technical roadmaps, mentoring engineers, and delivering measurable model improvements.

🏖️ Benefits

• Competitive compensation package including bonus components and stock option grants • Medical, dental, and vision coverage for full-time employees • Retirement savings plan (RRSP) with a 4% employer contribution • Public transit subsidy (Montreal region only) • Flexible working hours and generous paid vacation • Company-wide office closures during public holidays • Life insurance