AIML Developer

*Data Scientists don't apply*

Experience Requirement

• Minimum 5 years of professional experience in AI/ML development, with at least 2 years of hands-on experience in computer vision or video analytics.
  • Proven track record in developing and deploying real-time AI models for visual-based ITS or industrial automation products. • Experience working in production environments with large-scale data and multi-camera video analytics systems.

Key Responsibilities

• Design and develop AI-based video analytics models for incident detection, including stopped vehicles, wrong-way driving, pedestrian entry, congestion, and accidents.
• Develop deep learning solutions for vehicle classification, speed estimation, number plate recognition, and traffic rule violation detection.
• Build and optimize object detection, motion tracking, and event recognition pipelines for live video feeds.
• Integrate trained models with ITS frameworks such as VIDES, ATCC, AVC, TMCS, ANPR, and VASD.
• Develop tools for data labeling, annotation, and dataset management for model training and evaluation.
• Deploy AI models on edge computing platforms (e.g., NVIDIA Jetson, Intel Movidius) for real-time inference.
• Collaborate with cross-functional teams to integrate video analytics with control room dashboards, alerting systems, and central ATMS servers.
• Continuously monitor model performance and implement fine-tuning and retraining strategies to maintain accuracy.
• Collaborate with backend and frontend teams for API development, alert management, and visualization.
• Research and implement state-of-the-art techniques in computer vision and multimodal AI to enhance system capabilities.

Required Technical Skills

• Strong programming skills in Python, with experience in OpenCV, TensorFlow, PyTorch, or Keras.
• Proficiency in MySQL / Pinecone DB / Vector DB for model data integration and analytics reporting.
• Experience with REST API, GraphQL, GRPC, and integration of AI modules with backend systems using microservices architectures.
• Solid understanding of Deep Learning architectures CNN, R-CNN, YOLOv8, SSD, ResNet, and DeepSORT.
• Experience in real-time video processing, object detection, tracking, and event detection.
• Familiarity with RTSP, RTMP, ONVIF, and video streaming frameworks.
• Experience with edge AI deployment (TensorRT, ONNX, OpenVINO).
• Proficient in Linux environments, Docker, and Git version control.
• Skilled in model evaluation and performance tuning using metrics such as precision, recall, F1-score, and latency.
• Experience in computer vision dataset preparation, cleaning, and preprocessing.
• Expertise in messaging queues such as Apache Kafka, RabbitMQ, and AWS SQS.
• Expertise in data labeling, annotation, and building data pipelines for training/retraining ML and CV models.
• Proven experience in production deployment of AI models with minimal guidance.
• Ability to mentor junior team members in AI/ML concepts and best practices.

Preferred Skills

• Experience in ITS / Smart Highway / Smart City domains.
• Understanding of traffic enforcement rules and incident classification workflows.
• Familiarity with C#, .NET, Node.js, or React.js for backend integration.
• Experience with sensor fusion (Radar, LiDAR, and Video).
• Knowledge of MLOps practices for model lifecycle management and automated deployment.
• Experience with cloud-based AI inference platforms (Azure, AWS, or GCP).

Education

• Bachelors or Master’s degree in Computer Science, Artificial Intelligence, Electronics, or a related field.
  • Professional certifications in AI/ML, Computer Vision, or Deep Learning will be an advantage.

Location & Employment Type

Location: Noida
Employment Type: Full-time, On-site