SWE, Applied AI, and Security systems.

AI Enabled Monocular Gait Analysis

Clinical gait assessment system operating on standard single-camera video to evaluate neurological gait conditions (Cerebral Palsy, Stroke, Myopathic, Normal). Integrates YOLOv8 + ByteTrack tracking, 147-D pelvic-scaled kinematic pose features, and 28-D RAFT dense optical flow into dilated Temporal Convolutional Networks (TCN) reaching 70.7% balanced accuracy with SPARC smoothness profiling under strict video-grouped cross-validation.

Co-authored with Debeshi Sen (23BDS1055), guided by Dr. Pavithra L K, VIT Chennai

Weather-Based Traffic Signal Model

Adaptive traffic sign and signal control system combining real-time environmental/weather parameters with computer vision (EfficientNetB3) to dynamically adjust signal timings and maintain high recognition under adverse conditions (rain, fog, low visibility). Demonstrated an 86.2 percentage-point transfer advantage.

NABHA Health

Telemedicine platform for rural India built for Smart India Hackathon 2025. Voice-based AI intake via Twilio + Bolna, adaptive wristband health monitoring, doctor & pharmacy portals, and offline-first data sync.

NABHA Voice Intake

Voice-to-structured pipeline with retrieval-augmented summaries using Gemini API. Schema validation, confidence thresholds, and safe fallbacks ensure predictable UX for medical intake.