Good to have

YOLO, RT-DETR, Mask R-CNN, ViTPose, Keypoint R-CNN, ByteTrack, VideoMAE, Video Swin Transformer, MS-TCN, multimodal learning, synthetic data, domain adaptation, uncertainty estimation, or robotics applications

What you will do

- Model serial, redundant, cable-driven, and constrained robotic arms; develop forward/inverse kinematics, Jacobians, calibration, and redundancy resolution.

- Build constrained solvers for remote-centre, tool-axis, joint-limit, and workspace constraints.

- Generate and optimise smooth, precise, collision-aware trajectories; develop Cartesian, task-space, visual-servo, impedance, and admittance control.

- Integrate real-time execution, safety supervision, behaviour trees, recovery logic, and hardware interfaces.

- Establish deterministic baselines, compare solver and controller families,



analyse failure cases, and recommend an architecture. What we are looking for

- Strong C++ and Python skills with solid knowledge of robot kinematics, dynamics, Jacobians, and coordinate transformations

- . Experience implementing inverse kinematics, trajectory generation, robot controllers, constrained optimisation, or real-time numerical methods

- . Experience integrating simulators, sensors, actuators, and low-level robot interfaces, with measurable experimental validation

- . ROS 2 experience is helpful; expertise should extend beyond a single framework.

Work Location: In person

📌 Machine Learning & Computer Vision Engineer (Bengaluru)
🏢 Peoplevia Business Solutions
📍 Bengaluru