CronJobs

backend jobs

Software Engineer II - Fleet Enablement & Insights (Annotation Platform)

Torcrobotics · Ann Arbor, MI, Fort Worth, TX, Blacksburg, VA

hybridmid$139,000–$139,000Posted Aug 14, 2026TypeScriptReactthree.jsWebGLPythonSQLPostgreSQLAWS

Apply on the employer site

About this role

## Software Engineer II — Fleet Enablement & Insights (Annotation Platform) ### About Torc Torc believes 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 its solutions with experienced partners. Now part of the Daimler family, Torc focuses on developing software for automated trucks to transform how the world moves freight. ### Meet the Team You’ll join the **Fleet Enablement & Insights** team to build Torc’s in-house **annotation platform**—web-based tooling that turns multi-sensor autonomy data into labeled datasets used to train and validate the autonomous truck platform. You’ll develop an interactive **2D/3D annotation editor** and the services behind it, including fusing **HD map data** with **multi-camera, lidar, and other sensor context** into the labeling workflow. The platform also supports adjacent use cases such as **data QC** and **scene-selection review**. ### What You’ll Do - Design, develop, and maintain the **TypeScript/React** web application for 2D/3D annotation (cuboid/polygon editing, cross-frame interpolation & track propagation, attribute editing, and review-first workflows). - Build high-performance **point cloud and image rendering** with **three.js/WebGL** (octree/LOD streaming, predictive prefetching for smooth scrubbing, camera–LiDAR projection, multi-sensor overlays). - Build backend services behind the editor: label storage & versioning, task assignment & QA workflow, authentication, and multi-user isolation guardrails. - Integrate **pre-labeling/pseudo-labeling** outputs into the annotation workflow and instrument acceptance-rate/throughput metrics to improve the auto-labeling feedback loop. - Fuse **HD map data** into annotation and QC workflows as priors/reference layers. - Build data converters and ingestion paths from Torc multi-sensor scene data (multiple LiDARs, many cameras, calibration data) into platform formats. - Deliver dataset exports compatible with downstream ML training/validation, including lineage and auditability required for the safety case. - Use **AWS** services and **Databricks adjacency** to host scene data and deploy scalable services. - Collaborate with Data Annotation, Autonomy/ML, Scene Selection, Mapping, and Data Engineering to align with real annotator workflows and downstream requirements. - Participate in agile ceremonies, sprint planning, and weekly demos with annotators and stakeholders. - Contribute to engineering excellence via code reviews, documentation, and knowledge sharing; harden prototypes into production. - Identify and address technical debt, performance bottlenecks, and reliability gaps. ### What You’ll Need to Succeed - Strong proficiency in **TypeScript** and **React**, shipping production-quality web applications. - Experience with **browser-based 3D graphics** (three.js/WebGL or similar) or strong graphics fundamentals with the ability to ramp quickly. - Proficiency in **Python** for production backend services and APIs. - Experience working with **large sensor datasets** (point clouds, imagery, video) and optimizing data-heavy UIs. - Strong **SQL** skills and hands-on experience with **PostgreSQL** for label/task/metadata storage. - Experience with **AWS** for hosting data and deploying services. - Proficiency with **Git/GitHub** for version control and collaborative workflows. - Familiarity with **JIRA** (or similar) for t

Listing freshness

CronJobs last confirmed this listing 23h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.

Browse all software engineering jobs →

Follow fresh jobs in Discord