-- Streamlined cloud operations by developing a reusable AWS S3 module, enabling real-time file transfers and enhancing data accessibility.-- Optimized codebase to reduce space and time complexity, resulting in a 30% improvement in processing efficiency.-- Developed a robust log parser that transformed training logs into meaningful data, improving model observability and debugging.-- Implemented comprehensive unit, functional, and system-level testing, ensuring high reliability and… Show more -- Streamlined cloud operations by developing a reusable AWS S3 module, enabling real-time file transfers and enhancing data accessibility.-- Optimized codebase to reduce space and time complexity, resulting in a 30% improvement in processing efficiency.-- Developed a robust log parser that transformed training logs into meaningful data, improving model observability and debugging.-- Implemented comprehensive unit, functional, and system-level testing, ensuring high reliability and performance of the deployed models.-- Authored detailed technical documentation and code flow diagrams, enhancing team collaboration and onboarding efficiency.Impact: Enhanced the overall efficiency and reliability of deployment processes, significantly reducing resource usage and operational costs. Show less

Project: Foreign Particle Detection in Empty Water Bottle Cans on Production Line-- Collaborated with clients to understand requirements, leading to precise scope definition and successful project execution.-- Conducted R&D on various industrial-grade cameras, optimizing image capture quality for improved detection accuracy.-- Trained multiple deep learning networks (CNN, SSD, YOLO v8, FASTER RCNN) on cloud GPU, achieving over 95% detection accuracy.-- Deployed models using… Show more Project: Foreign Particle Detection in Empty Water Bottle Cans on Production Line-- Collaborated with clients to understand requirements, leading to precise scope definition and successful project execution.-- Conducted R&D on various industrial-grade cameras, optimizing image capture quality for improved detection accuracy.-- Trained multiple deep learning networks (CNN, SSD, YOLO v8, FASTER RCNN) on cloud GPU, achieving over 95% detection accuracy.-- Deployed models using Docker, ensuring seamless integration and scalability.Led onsite deployment in Dubai, successfully integrating the system into the client’s production line.Impact: Reduced foreign particle contamination by 80%, enhancing product quality and client satisfaction.Project: Conversational AI for Sales & Marketing -- Fetched user profile data using LinkedIn API, enhancing data-driven insights.-- Performed sentiment analysis on comments, providing valuable feedback during corporate meetings.Impact: Developed a prototype for real-time sentiment analysis, improving meeting productivity and decision-making.Project: Pothole Detection System-- Gathered and annotated extensive image datasets, ensuring comprehensive training data.-- Optimized neural network models for edge devices, facilitating real-time detection on smartphones and Raspberry Pi.Impact: Developed a cost-effective solution for pothole detection, improving road maintenance efficiency.Project: Surface Defect Detection in Iron Bars-- Visited client location to understand the problem use case and gather domain knowledge.-- Developed deep learning models in collaboration with SMEs, detecting defects with over 90% accuracy.-- Created visualization charts, providing actionable insights on defect types and frequencies.Impact: Improved defect detection processes, reducing manual inspection time by 60% and increasing operational efficiency. Show less