• Building Python and Django REST APIs to support GenAI-powered internal tools, focusing on reducing API response latency across critical business endpoints.

• Contributing to LangChain-based LLM pipelines integrated with vector databases and FastAPI for financial reporting workflows.

• Supporting cloud-native deployments on AWS using Docker and Kubernetes, with monitoring via CloudWatch and for LLM observability.

• Implementing prompt engineering strategies including chain-of-thought reasoning and few-shot learning to optimize token usage and output quality.

• Assisted in enhancing existing LangChain-based LLM pipelines by fine-tuning prompts and improving retrieval logic, leading to better response relevance in financial reporting use cases.

• Supported debugging and performance optimization of Django and FastAPI services, identifying bottlenecks and contributing to reduced API latency across key endpoints.

• Collaborated with cross-functional teams to test and validate GenAI outputs, ensuring alignment with business requirements and improving overall reliability of AI-driven workflows.

• Designed and implemented scalable backend using RESTful APIs in Python (Django) to support high-volume, low-latency retail data workflows such as orders, events, and customer interactions.

• Built and exposed Node.js-based REST APIs to enable seamless integration between frontend applications and backend services for real-time data exchange.

• Developed API-driven services for ingesting, validating, and persisting transactional and behavioral data into MongoDB, PostgreSQL, and DynamoDB

 Built and deployed Generative AI (GenAI) integrations leveraging modern frameworks and APIs to support automation, content generation, and NLP-driven workflows.

 Developed high-performance RESTful APIs and microservices to serve AI/ML models and GenAI functionalities in production environments.

 Engineered data processing and feature pipelines to support machine learning use cases, ensuring data quality, reliability, and scalability.

 Collaborated with cross-functional teams to integrate AI/ML and GenAI capabilities into business applications, enhancing user experience and system intelligence.

 Deployed cloud-native AI services using containerization, CI/CD pipelines, and DevOps best practices, improving system resilience and release efficiency.

 Managed the full software development lifecycle (SDLC) from concept to deployment, utilizing Git for version control and implementing CI/CD pipelines to automate testing and release processes.

 Optimized application performance by profiling Python code and database queries, resulting in a $30\%$ improvement in API response times for critical business functions.

• Implemented automated unit, integration, and end-to-end tests using frameworks like Pytest and Django's built-in testing tools, leading to a 45% reduction in post-release bugs.

• Successfully containerized application components using Docker for development and staging

environments, streamlining the onboarding process for new team members.

• Built AI chatbots using GPT-4. LangChain, and vector databases to automate customer support workflows, reducing ticket volume by 6o% while maintaining high satisfaction scores.

• Created internal Al assistants using LangChain agents and LLMs to automate code, documentation, and knowledge retrieval tasks, increasing developer productivity by 30%

• Designed and deployed LLM-powered applications using OpenAl GPT-4› LLaMA, and Mistral with LangChain for orchestration, memory, and tool calling, reducing manual research and analysis effort by 60%

• Built Retrieval-Augmented Generation (RAG) systems using LangChain retrievers,

FAISS/Pinecone/ChromaDB, and Hugging Face embeddings, improving response accuracy and

factual grounding by 42%

• Developed Python-based ETL pipelines to ingest, cleanse, and transform clinical trial datasets from HL7 FHIR and EDC systems, processing 500,000+ patient records with 99.8% data accuracy for downstream analytics.

• Built RESTful microservices using FastAPI and Flask to serve AI-driven insights for clinical data

management, reducing manual data review cycles by 35% for life sciences clients.

• Assisted in designing NLP modules using Hugging Face Transformers and spaCy for automated extraction of adverse events and medical entities from unstructured clinical narratives, improving annotation throughput by 50%.

• Developed responsive front-end dashboards using React, Redux, and Material UI to visualize patient cohort analytics and trial metrics, enabling real-time decision-making for clinical operations teams.

• Integrated backend APIs with PostgreSQL and MongoDB databases using SQLAlchemy ORM and Celery task queues, ensuring reliable asynchronous data processing for high-volume healthcare data workflows.

• Collaborated on HIPAA-compliant cloud deployments on AWS using S3, Lambda, and CloudWatch, ensuring secure handling of protected health information (PHI) across all development and staging environments.

• Participated in Agile sprint ceremonies and peer code reviews, refactoring legacy Python scripts into modular, well-documented packages, improving code maintainability and reducing technical debt by 20%.

• Developed REST API consumers using Python GET and POST methods, validated with Postman, reducing integration defects by 30% during sprint cycles.

• Built and maintained data parsing modules to process CSV, XML, and JSON from REST services, automating ingestion workflows and saving 10+ hours of manual effort weekly.

• Designed interactive web-based solutions using Python, Django Forms, HTML5, and JavaScript, enabling real-time data collection for 20,000+ online users.

• Wrote Python modules to extract and load asset data from MySQL source databases, reducing manual data migration effort by 100% for 3 legacy system integrations.

• Participated in full Agile/Scrum ceremonies including daily stand-ups, sprint planning, and retrospectives, contributing to on-time delivery across 10 consecutive sprint cycles.