M

Mecka AI

Labeled human movement data for AI

Senior SLAM & Calibration Engineer

Robotics & Autonomous SystemsConfirmed live in the last 24 hours
Full-timeCA$200k - CA$230k /yr

About the Job

Requirements

  • At least 5 years of hands-on experience building, maintaining, or improving SLAM, visual-inertial odometry, and Structure-from-Motion pipelines, including production systems.
  • Hands-on experience with online or continuous calibration and camera and IMU calibration, including extrinsics, intrinsics and their limits, IMU noise density and random-walk bias, lens-distortion models, and temporal synchronization.
  • Ability to own calibration across noisy real-world devices and build monitoring that detects drift.
  • Understanding of the linear algebra, optimization, and first principles behind spatial tracking.
  • Ability to write clean, efficient, scalable C++ and Python.
  • Ability to specify calibration requirements clearly to a hardware team across time zones.

Responsibilities

  • Develop and maintain continuous, in-pipeline camera calibration that refines extrinsics from recorded, hardware-synced, uncompressed video and IMU, enabling fixes to ship without a device software update.
  • Fix and harden the fleet-wide video-based calibration refinement system, including its monocular wrist-camera path, and validate it against real benchmarks.
  • Build monitoring and metrics tracking to detect calibration drift across devices, flag devices for recall, and maintain the calibration database and version history.
  • Maintain and improve the SLAM and visual-inertial odometry systems supporting online calibration, and emit calibration-related error as a first-class output.
  • Coordinate with the factory-calibration engineer and China hardware team on intrinsics limits, IMU noise and bias parameters, temporal synchronization, and validation runs that avoid local minima.
  • Ship interactive Rerun and Gradio tools to visualize trajectories, drift over time, reprojection error, and per-device calibration metrics for operations and researchers.

Desired Qualifications

  • Experience with video-based or markerless calibration and refinement systems.
  • Experience with sensor-fusion and calibration tools such as Kalibr, GTSAM, or Ceres Solver.
  • Experience with OpenCV, COLMAP, PyTorch, and FFmpeg.
  • Experience monitoring at scale, including calibration databases and versioning.
  • Experience with Rerun, Gradio dashboards, or trajectory and dataset browsers.
  • Experience with machine-learning infrastructure or data pipelines that operate at scale.

About the Company

Mecka AI

Labeled human movement data for AI

Mecka AI builds custom, labeled datasets of human movement by recording people performing specific tasks and analyzing the video to create multimodal data for AI and robotics training. Its offerings are data-as-a-service: bespoke, high-quality video data that is anonymized, meticulously annotated, and ready for direct ingestion into machine learning models. The company differentiates itself with a global, on-demand data-collection network and task-focused labeling to supply precise, task-specific datasets for autonomous systems, including humanoid robotics partners. Its goal is to accelerate the development and deployment of embodied AI and autonomous technologies by providing scalable, ready-to-use movement data.

Company size
51-200
Company stage
Series B
Total funding
$120M
Headquarters
New York City, New York
Founded
2025

Recently Posted Jobs

Explore more opportunities at Mecka AI

See more jobs