• Developed backend inventory and ordering workflows using Node.js and Express by enhancing transaction handling and report calculation logic, reducing financial and operational data inconsistencies by 30% .

• Built an automated invoice processing pipeline by integrating external APIs to fetch invoice data, converting files into PDF format, and storing them in AWS S3, eliminating manual upload efforts and improving operational efficiency.

• Architected and optimized serverless ETL workflows using AWS Lambda, implementing structured logging, error handling, and monitoring (CloudWatch), increasing system reliability and simplifying debugging by 43%.

• Reduced technical debt by refactoring legacy API client code into reusable HTTP abstractions and building backend utilities to transform complex pricing and catalog data into structured models, improving maintainability and frontend efficiency by 67%.

• Improved code quality and system reliability by implementing unit and integration testing, leading code reviews, and enforcing development best practices across end-to-end services.

• Developed scalable backend services using Java and Spring Boot for and event-driven architectures, improving service request processing efficiency by 17% and reducing operational costs by $20K annually.

• Built asynchronous processing pipelines using multithreading and messaging systems (Kafka/RabbitMQ), increasing system throughput by 25% and reducing latency during peak traffic.

• Optimized database performance using SQL indexing, query tuning, and schema design improvements, enhancing query efficiency and overall application performance.

• Designed and implemented RESTful APIs to enable seamless communication across distributed systems, improving data consistency and reducing integration issues.

• Owned production issue resolution by leading debugging and root cause analysis, improving system stability and reducing recurring defects, while mentoring junior developers and conducting code reviews.

• Built a RAG-based healthcare document analysis system using Python, FastAPI, LangChain, and FAISS to process clinical notes, discharge summaries, and lab reports, reducing manual review effort by 19%.

• Engineered robust document ingestion pipelines for structured and unstructured medical data (PDF/text), including text extraction, chunking, metadata tagging, and embedding generation to enable efficient semantic search at scale.

• Implemented contextual question-answering by integrating OpenAI embeddings with FAISS vector store, improving retrieval relevance by 25% compared to traditional keyword-based approaches.

• Developed REST APIs for document upload, retrieval, and summarization, enabling near real-time extraction of key entities such as diagnoses, medications, and follow-up instructions.

• Enhanced system reliability and response quality by building evaluation pipelines, implementing retrieval validation and grounding checks, and ensuring data privacy through de-identification aligned with healthcare compliance standards.