Job Description

We are looking for highly skilled Computer Vision Architects / Engineers with strong expertise in deep learning, real-time video analytics, and edge AI deployments. The ideal candidate will have hands-on experience in building, optimizing, and deploying computer vision and AI/ML models at scale, particularly for image and video-based use cases using Python, PyTorch, TensorFlow, and NVIDIA platforms.

This role involves working on large-scale, real-time systems (100+ video streams) and contributing to end‑to‑end AI pipelines—from model development to edge & cloud deployment.

Key Responsibilities-

Design, train, optimize, and deploy computer vision models for:
Object detection & recognition
Image classification & segmentation
Action recognition and video analytics
Lead and contribute to large-scale real-time deployments handling 100+ concurrent video streams.
Fine-tune and optimize CNNs, Vision Transformers (ViT) and other deep learning architectures.
Build scalable inference pipelines using PyTorch, TensorFlow, ONNX.
Deploy and optimize models using NVIDIA AI stack, including:
NVIDIA TAO Toolkit
DeepStream SDK (Metropolis)
Triton Inference Server
NGC Pre-trained Models
Implement advanced multi-object tracking (MOT) across multi-camera systems.
Perform camera calibration, ROI extraction, and spatial-temporal analysis.
Develop feature extraction models, latent space analysis strategies, and representational learning techniques.
Apply robust pre-processing, post-processing, and evaluation techniques to improve model accuracy and performance.
Collaborate with cross-functional teams on MLOps, CI/CD pipelines, and edge deployment strategies.