Lavanya Vijayakrishnan

• Designed and developed a serverless trading platform for recyclable goods with event-driven architecture, enhancing system reliability and scalability for uninterrupted trading. Implemented an asynchronous notification module, reducing API response time by 15 percent, thereby boosting overall system performance. • Architected and developed an advanced software solution for land parcel value estimation, prominently employing object-oriented design patterns to ensure robustness and scalability. This system integrates a machine learning module for precise land information analysis. The core functionality includes complex data processing calculations, culminating in a visually intuitive interface for user-centric presentation of land parcel valuations.

• Designed and implemented an ETL service using AWS Glue, incorporating a robust data validation module. This initiative resulted in a 15 percent decrease in data duplication and mismatch incidents, significantly enhancing data integrity through comprehensive validation procedures. • Efficiently led the migration of legacy software applications to a modern, scalable microservice, including remodelling the database schema and transitioning to NoSQL solutions, ensuring zero data loss; additionally, my architectural design inputs have optimized the system efficiency by at least 20 percent with minimal downtime, and received praise for swift, error-free execution. • Developed a custom machine learning model using Azure Form Recognizer for streamlined data extraction from surveys and trained an image captioning model. Integrated both models into a Django application as RESTful APIs. Identified CPU bottlenecks and incorporated different optimization techniques there by improving throughput by over 20 percent. Also created a Dialogflow chatbot for efficient car service booking and customer interaction.