Results-driven Graduate Trainee Engineer with expertise in data science and computer vision at Go Digit General Insurance. Successfully implemented advanced models for car damage detection and developed a robust Aadhaar face verification system. Proficient in Docker, Kubernetes, and Python, with strong analytical skills and a commitment to delivering innovative solutions. Resourceful Graduate Trainee Engineer known for high productivity and efficient task completion. Possess specialized skills in project management, process optimization, and technical troubleshooting. Excel in problem-solving, teamwork, and adaptability, ensuring seamless collaboration and innovative solutions in engineering projects.
Car Damage Detection (PI & Claims)
Trained separate YOLO models for pre-inspection (scratches, dents) and claims (bumper damage, cracks)
- Achieved 90%+ accuracy in real-world testing.
Odometer Reading Detection
- Used YOLOv11 to detect odometer regions (87.1% accuracy) and extracted readings with OCR.
- Integrated into insurance inspection workflows.
4-Wheeler Pose Detection (PI & Claims)
- Classified 8 vehicle orientations for pose validation.
Achieved 91.27% accuracy and improved damage localization quality
Aadhaar Face Match System
- Verified selfie–Aadhaar matches using Glintr100 embeddings + Siamese classifier.
- Reached 85.7% accuracy and designed to flag potential fraud.