> Markdown version of [/jobs/ext/4749767-computer-vision-engineer-3d-on-site](https://www.wearedevelopers.com/jobs/ext/4749767-computer-vision-engineer-3d-on-site). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Computer Vision Engineer, 3D - ON-SITE - **Company:** Targeted Talent - **Location:** New York, NY, United States - **Experience:** Expert - **Salary:** $200,000 - $230,000 - **Contract:** Permanent contract - **Skills:** Computer Vision, C++ (Programming Language), Python (Programming Language), Kinematics, Signal Processing, Software Engineering, Scripting - **Published:** November 16, 2026 - **Apply:** https://www.careerbuilder.com/job-details/computer-vision-engineer-3d-on-site-new-york-ny--f6d224d4-481e-4d42-86f9-171bd32f5386 ## About the Role Computer Vision, Computer Vision, Geometry, Calibration, Perform Arrests, Quality Metrics, Reporting Dashboards, Software Engineering, Python Programming/Scripting Language, C++ Programming Language, * 5+ years of professional software engineering and computer vision experience * Strong, coding-heavy production Python (C++ a plus) * Solid multi-view geometry: triangulation, PnP, calibration, reprojection, coordinate transforms * Temporal filtering and signal processing, with a feel for smoothness versus fidelity * Inverse kinematics and rigid-body or bone-length constraints on articulated structures * Enough ML to work with pose and mesh regressors and their failure modes * A data-quality mindset with noisy real-world data Nice to have * Hand or body pose and mesh reconstruction * Bayesian filtering or state estimation; Ceres or GTSAM * CV tooling: trajectory visualizers, overlay or QA dashboards * Stereo, multi-view or ego-exo capture; light temporal ML for smoothing or gap-filling ## Description Own the classical side of the computer vision stack. Deep models produce raw per-frame predictions; you'll own the traditional CV, temporal filtering and inverse kinematics that turn them into clean, smooth, correctly calibrated 3D data ready for delivery. You'll partner closely with research teams and need to understand the models, but you won't train them. The immediate focus is a 3D hand-tracking pipeline, then classical-CV problems across the organization. What you'll do * Own multi-view geometry: triangulation, PnP, camera calibration, undistortion and reprojection * Design and tune temporal filters (low-pass, Kalman-family, interpolation) that remove jitter without losing real motion * Run inverse kinematics and enforce physical constraints such as bone-length rigidity * Build QA checks and metrics that keep bad frames out of customer deliveries, and keep the pipeline running at scale * Build visualizers and dashboards for trajectories, overlays and failure cases