Computer Vision Scientist-AI Labs

Purpose/Objective

The Computer Vision Scientist will be responsible for developing, training, and deploying AI-powered computer vision models to enhance automation, efficiency, and decision-making across industries. This role focuses on model development, data processing, and real-time implementation, working closely with cross-functional teams to ensure high-performance computer vision algorithms are effectively integrated into business operations.

Key Responsibilities of Role

Computer Vision Scientist-AI Labs AI Model Development & Deployment: Design, train, and optimize deep learning-based computer vision models, including Object Detection (OD), Optical Character Recognition (OCR), 3D vision, Event Detection (ED), Satellite Image Processing (SI), and Thermal Imaging (TI). Develop custom deep learning architectures using CNNs, transformers, GANs, and YOLO-based frameworks for real-time applications. Implement and fine-tune RGB-D image processing algorithms for depth estimation, multi-view stereo, and LiDAR-based vision. Optimize AI inference speed, accuracy, and computational efficiency to support real-time processing at the edge and cloud environments. Data Engineering & Feature Extraction: Preprocess large-scale image video datasets, performing data augmentation, feature extraction, and dimensionality reduction. Develop custom image segmentation, object tracking, and feature detection pipelines. Implement efficient data pipelines for structured and unstructured image processing, integrating with HPC clusters and cloud-based storage. AI Training & Model Optimization: Train state-of-the-art neural networks (CNNs, RNNs, ViTs, RAG, GANs, Transformers, etc.) on complex datasets. Implement MLOps workflows, leveraging TensorFlow, PyTorch, OpenCV, and Hugging Face for automated model training. Ensure model generalization by experimenting with data augmentation, fine-tuning, and adversarial training techniques. Image Analytics & Real-Time AI Deployment: Deploy AI models on cloud (AWS, GCP, Azure) and edge devices (Jetson Nano, TensorRT, OpenVINO). Integrate computer vision models with IoT cameras, satellite feeds, and industrial automation systems. Develop real-time analytics dashboards to visualize AI model insights for business stakeholders. Cross-Functional Collaboration & Execution: Work closely with AI engineers, data scientists, and software teams to integrate CV models into enterprise applications. Support R&D teams in testing and refining AI-powered robotics, drone vision, and autonomous navigation solutions. Collaborate with business teams to align AI-driven insights with operational goals and decision-making. AI Governance, Compliance & Risk Management: Ensure model fairness, bias mitigation, and explainability for AI-driven computer vision applications. Comply with data privacy regulations (GDPR, AI Act) and ethical AI standards for responsible AI development. Monitor and address AI model drift, hallucinations, and security vulnerabilities. Key Stakeholders - Internal Computer Vision Head Data Science & Engineering Teams Business & Product Teams Cloud & DevOps Teams Key Stakeholders - External AI Research Institutions & Tech Partners Regulatory Bodies Cloud & AI Infrastructure Providers

Technical Competencies

AI Governance, Compliance & Ethical AI-SDIL,AI Model Development & Optimization-SDIL,AI Product & Solution Development-SDIL,AI Research, Experimentation & Emerging Technologies-SDIL,AI Strategy & Enterprise Transformation-SDIL,Cross-Functional Collaboration & AI Integration-SDIL

Qualifications and Experience

Educational Qualification: Master’s Ph.D. in AI, Computer Vision, Machine Learning, or Computer Science (Preferred from ISI IISc IIT Top AI Institutes) Certification: (Preferred but Not Mandatory) AWS GCP Azure AI & Machine Learning Certification Deep Learning Specialization (Coursera, Stanford, Andrew Ng’s DL Program) Computer Vision Nanodegree (Udacity, MIT AI) DevOps for AI & MLOps Certification (Google, Kubernetes, TensorFlow Extended – TFX) Work Experience (Range of years): 5-10 yrs