Overview
We are seeking a highly skilled Computer Vision Engineer specialising in real- time detection, pose estimation, and movement analysis to design and deliver robust vision- based intelligence systems. The role focuses on building scalable pipelines capable of detecting people and objects, analysing human movement patterns, and tracking individuals across frames in real-world environments.
This position will play a critical role in developing production-grade systems that operate reliably across varying lighting conditions, camera angles, occlusions, and live video feeds, ensuring consistent performance in real deployment scenarios.

Key Responsibilities
• Design and implement real-time computer vision pipelines for person and object detection.
• Develop advanced pose estimation and movement analysis models to interpret body
posture, gestures, and motion patterns.
• Build multi-object tracking systems to accurately track individuals across frames and
maintain identity persistence over time.
• Optimise models for low-latency inference suitable for edge and near-edge deployments.
• Ensure system robustness across diverse environmental conditions, including varying
lighting, occlusions, and camera perspectives.
• Integrate computer vision models with video ingestion, streaming, and processing
frameworks.
• Perform model evaluation and benchmarking using real-world datasets and live feeds.
• Collaborate with cross-functional teams to align vision models with business use cases
and operational requirements.




• Continuously improve detection accuracy and tracking stability using state-of-the-art
techniques.

Skills & Qualifications
Strong expertise in Computer Vision and Deep Learning techniques is required. The candidate should have hands-on experience with object detection approaches such as YOLO,Faster R-CNN, and SSD; pose estimation frameworks such as OpenPose, HRNet, and MediaPipe; and multi-object tracking methods such as DeepSORT, ByteTrack, and OC-SORT.

The role requires proficiency in Python and deep learning frameworks such as PyTorch or TensorFlow, along with experience in real-time video processing pipelines using OpenCV, GStreamer, or FFmpeg.

A strong understanding of image processing, feature extraction, model optimisation, GPU acceleration, and performance tuning is significant for delivering reliable systems in production environments.

Experience in edge deployment and inference optimisation using TensorRT, ONNX, and
quantisation techniques will be valuable. The candidate should also be comfortable evaluating models using precision, recall, tracking accuracy, latency, and related metrics while balancing trade-offs between accuracy, performance, and cost.

Experience : 5+ years of experience in computer vision engineering or applied AI development, with a proven track record of building and deploying real-time vision systems in production environments. The candidate should have practical experience working with live video feeds, real-world datasets, and deployment scenarios where lighting, camera angles, occlusions, and operating conditions may vary significantly.

📌 ML Engineer - Computer Vision (Jaipur)
🏢 Durapid Technologies
📍 Jaipur