AI/ML Developer (Toll industry)

About The Opportunity A fast-scaling tech firm operating at the intersection of AI, transportation infrastructure, and smart tolling systems, we build next-gen machine learning models that power real-time vehicle classification, license plate recognition, fraud detection, and dynamic toll pricing for national highway networks across India. Our AI/ML developers design and deploy computer vision and NLP-driven solutions that process millions of toll transactions daily — turning raw camera feeds and sensor data into intelligent, automated, and scalable mobility infrastructure.

Role & Responsibilities

• Design, train, and deploy deep learning models for vehicle type classification, license plate OCR, and occupancy detection using CV pipelines.
• Optimize inference latency and accuracy for edge-based toll gate cameras and cloud-based central processing units.
• Integrate ML models into production-grade tolling platforms using Docker, Kubernetes, and MLOps tools like MLflow or Weights & Biases.
• Collaborate with hardware engineers to calibrate ML performance under varying lighting, weather, and traffic conditions.
• Build anomaly detection models to flag toll evasion, duplicate transactions, or device tampering in real-time.
• Document model performance metrics, versioning, and retraining workflows to ensure regulatory and operational compliance.

Skills & Qualifications

• Must-Have
• Python
• PyTorch
• TensorFlow
• OpenCV
• YOLOv8
• ONNX
• Docker
• REST APIs
• Preferred
• ONNX Runtime
• TensorRT
• Tolling or transportation domain exposure

Benefits & Culture Highlights

• Work at the forefront of AI in India’s smart infrastructure ecosystem — your models impact national highways and millions of daily commuters.
• Access to real-world datasets, edge hardware labs, and cloud GPU clusters for rapid experimentation and deployment.
• Flat hierarchies, outcome-driven culture, and direct exposure to product owners and government stakeholders. Skills: learning,models,tolling,classification,docker,edge,design,cloud,ml,infrastructure