Jan 2023 – Oct 2025 · 2 yrs 9 mos · Pune, Maharashtra, India · On-site

• Led the end-to-end design and development of a production-grade AI Assistant microservice for Oracle Primavera Cloud leveraging RAG, intelligent chunking strategy, multi-tool AI agent support with multiple data sources integration. This reduced help search time from 10 minutes to 3 seconds and projected to enhance client retention by 15-20%.
• Optimized a Python FastAPI microservice performance by implementing asyncio-based parallel processing, slashing integration test execution time by 78% and enabling high concurrent user handling via thread pool offloading.
• Optimized LLM outputs with context-aware prompt generation, improving platform support accuracy and reducing manual intervention.
• Optimized page load performance for one of the modules of Oracle Primavera Cloud by implementing lazy loading, partial loading and other advanced optimization techniques, achieving an 83% improvement.
• Architectured and developed a cloud-based tender marketplace app with Oracle APEX, achieving 20x speed and 100x code reduction.
• Engineered and integrated email, eLocation, and file-sharing microservices within the application, utilizing NGINX load balancing to enhance scalability, reliability, and performance under heavy traffic.
• Developed an internal utility tool that provides detailed and precise analytics on data modifications across multiple schemas, triggered by operational activities, enhancing developer productivity and efficiency.
• Developed a robust bidding platform using Oracle APEX, streamlining contract management, and enhancing user experience.
• Utilized Agile/SCRUM, executing biweekly sprints for timely updates & seamless collaboration with a 7-member developer team.
• Leveraged SQL, PL/SQL, JavaScript, HTML, and CSS to create an intuitive and responsive web application.

Dec 2021 – Jul 2022 · 7 mos · India · Remote

• Automated Data fetching from the API, streamlining workflows which saved 10,000 hours of developer's work.
• Added a feature for automatic data sharing which reduced the cases of theft by almost 100%.
• Developed a full fledged data capturing app with image processing features which is used by the computer vision team.
• Migrated the application from a file-based storage system to Room SQL Database, significantly enhancing data management and stability. This transition led to a 99% crash-free experience and improved overall app reliability.
• Redefined the app’s code according to MVVM architecture which makes automation and scaling easier.
• Integrated native C++ code for image processing in Android using JNI, boosting app speed by 25% and enhancing security, making reverse engineering nearly impossible and significantly improving user experience.
• Changed all the deprecated libraries like AsyncTask to Worker and Executor which prevent unnecessary memory leaks.
• Developed a data capturing app with image processing features, reducing manual effort by 40% and boosting system performance by 30% for faster project execution by the computer vision team.

May 2021 – Jul 2021 · 2 mos · Silchar, Assam, India · On-site

• During my summer internship, I contributed to a critical digital healthcare project focused on the early detection of Diabetic Foot Ulcers (DFUs). I developed and trained a deep learning model using the EfficientNet architecture to automatically classify DFU images. By leveraging a model that intelligently scales network depth, width, and resolution, my solution achieved an outstanding accuracy of 98.9%, significantly outperforming other established models and demonstrating the potential of AI to improve patient outcomes.