Construction · Remote

Senior, ML Engineer - 3D Reconstruction

Torc Robotics · Remote · Posted Nov 25, 2025

This posting comes from Torc Robotics. Confirm the details on their site before applying.

Overview

The Pseudo-Labeling team's goal is to create high-quality annotations on sensor data (images, point clouds). The annotations include 2D, 3D bounding boxes, classes, trajectories, lane lines, segmentations, depths, and high-definition map elements. The annotations are then used by different downstream users — for example, perception teams use them to train various models, mapping teams use them to build and maintain HD maps, and simulation teams use them for generating new data.

Responsibilities

  • Own the design, implementation, and deployment of offline 3D reconstruction, lane line detection, and automatic mapping/map creation systems that generate high-quality annotations on Cloud Services from logged sensor data (Cameras, Lidars, Radars, GPS/IMU).
  • Lead lane line network reconstruction efforts, rebuilding topologically correct lane line graphs from noisy, partial, or conflicting detections across drives.
  • Design, train, and iterate on lane line detection and creation models, pushing state-of-the-art performance and driving model architecture and training strategy decisions for the team.
  • Develop methods for creating lane line maps in previously unmapped areas, and for detecting and propagating map updates as road geometry changes over time.
  • Own pose correction for sensor data, and lead the diagnosis and mitigation of calibration and sensor errors (extrinsic/intrinsic miscalibration, drift, noise) that degrade reconstruction and mapping quality.
  • Set technical direction and best practices for the team's lane line, mapping, and pose estimation pipelines, and lead architecture decisions with cross-team impact.
  • Serve as technical lead on high-complexity projects, guiding and mentoring other engineers across multiple facets of project execution.
  • Stay ahead of the latest developments in AI and ML for autonomous driving, 3D reconstruction, and automated mapping, and bring emerging techniques into production.
  • Independently develop offline perception and mapping models or algorithms using disciplined software development processes, making recommendations for developing new code or re-using existing code, implementing version control, and maintaining documentation of created applications.
  • Define and implement ingestion, data preparation, curation, and governance of large, multi-faceted data sets supporting analytics models and workflows.
  • Proactively assess current capabilities to identify areas for improvement, proposing solutions that align with core strategy and operation.
  • Measure and track auto-labeling and map creation quality to meet internal customer requirements, including quantifying the impact of pose and calibration error on downstream quality.
  • Guide and produce information products, supporting visualization and data accessibility in a customer-centric manner.
  • Evaluate and make recommendations regarding technical advances that improve productivity and quality, reduce flow times, and enhance operational surety.
  • Develop guidelines and standards for analytics and machine learning models, their deployment, and associated processes.

Requirements

  • Considered an expert and highly skilled in discipline; conducts highly complex, important work independently with minimal to no supervision and broad latitude for independent judgment.
  • Scope of Influence: Expected to drive alignment across team and organizational boundaries. Sets technical direction, owns team technical solutions end-to-end, and is looked to as the authority for lane line, mapping, and pose estimation problems.
  • Mentors and guides engineers within and beyond the group.
  • Bachelor’s Degree in Computer Science, Robotics, Electrical Engineering or related technical field plus demonstrated competencies and technical proficiencies typically acquired through 6+ years of experience OR; Master’s Degree in Computer Science, Robotics, Electrical Engineering or related…
  • Strong, demonstrated background in lane line detection and creation, including hands-on experience with lane line network reconstruction and training lane line creation models.
  • Proven experience creating lane line maps for previously unmapped areas, and handling ongoing map updates as road conditions and geometry change.
  • Strong background in pose correction, with direct experience diagnosing and correcting calibration and sensor errors across camera, lidar, radar, and GPS/IMU systems.
  • Active Learning & Pseudo-labeling – Computer Vision, Deep Learning, Model training at an advanced level.
  • Two or more of the following at depth: 2D/3D Object Detection, Tracking, Sensor Fusion, Semantic Segmentation, SLAM, BEV.
  • Scaled ML Operations (MLOps) and Tooling – ML Frameworks, experiment tracking, model registry, MLflow, Weights and Biases, ML Metrics and Evaluation / Quality.
  • Distributed machine learning frameworks – PyTorch, Lightning, Ray.
  • Model Data Curation – Parquet data processing (PyArrow, Daft, Pandas, etc).
  • Development Tools & Eco-System (at scale) – Advanced proficiency in Python software development. Also, VDI and cloud-based development environments, CI Systems (GitHub Actions), and Docker.

Preferred qualifications

  • PPK/RTK (Post-Processed Kinematic / Real-Time Kinematic) GPS experience.
  • GIS (Geographic Information Systems) experience.

Benefits

  • A competitive compensation package that includes a bonus component and stock options
  • 100% paid medical, dental, and vision premiums for full-time employees
  • 401K plan with a 6% employer match
  • Flexibility in schedule and generous paid vacation (available immediately after start date)
  • AD+D and Life Insurance

Great fits start here.

Whether you're looking for your next role or your next team member, give people and possibilities a place to meet.