Engineered advanced computer vision models using deep learning frameworks to extract actionable insights and drive automation. Designed and implemented a microservices-based dashboard that achieved sub-second response times, significantly enhancing system scalability. Led cross-functional initiatives to integrate AI-driven solutions into production pipelines, boosting overall operational efficiency. Pioneered research and development in real-time image processing, establishing new performance benchmarks within the organization. Developed a real-time mosquito tracking and counting program that analyzes video streams using optimized mathematical models and critical reasoning—achieving high performance without GPU acceleration.

Architected an AWS IoT solution integrating IoT Core, Lambda, RDS, and S3 for seamless device onboarding and OTA updates. Developed a computer vision system for hotspot/coldspot detection using a U-Net model with a React/Tailwind CSS frontend. Engineered a Raspberry Pi–based telematics solution integrating 4G, GPS, and OBD-II for real-time vehicle analytics. Implemented a Samsung SmartThings Edge Driver with custom webhook integration, enhancing device capabilities.

Led a team of 3 in developing a dispatch simulator project utilizing Long Short-Term Memory (LSTM) models. Coordinated cross-functional efforts to meet key project milestones and deliver a high-quality solution.

Integrated a linter that resolved 3,000+ bugs, enhancing app stability and enabling continuous code quality checks. Fixed dependency conflicts (e.g., jitsi_meet and stripe_sdk) to ensure successful builds on Android 12. Solved a UI bug where expanded tiles lost state during video playback by storing state in local storage, improving user continuity. Introduced a custom Vimeo video player UI to enhance video playback quality. Incorporated update notifications for Play Store releases, ensuring the app remained current. Reduced app size by 20 MB through the elimination of unused dependencies.

Optimized LMS app performance through advanced data parallelization, slashing load times from 6 seconds to milliseconds. Developed a full-stack CRM for property management using React, Flutter, Firebase, Django, and Twilio. Created a custom BLE library for iOS, doubling command throughput in a React Native robot controller app. Spearheaded AWS Amplify integration for an educational app, achieving 5,000+ downloads on the Play Store.