Software Engineer & Researcher
A self motivated software engineer with over 7 years of experience, specializing in Data Science and Artificial Intelligence. Driven by a passion for data, my current R&D focuses on Large Language Models, vector databases, and model fine-tuning to build tailored AI applications that solve real-world organizational challenges.
More About MeSystem to verify borrower records (PII, income, application status, denial reasons etc.) by cross-referencing multiple source documents. Outputs exact citations/page numbers, confidence metrics, and concise reasoning for auditability and fair-lending compliance.
Secure, proprietary AI chat application for internal employees in compliance-restricted environments. Includes safe web search, deep-research capabilities, and a sandboxed code agent that executes code and returns structured outputs and visualizations. Eliminates reliance on external AI providers while maintaining enterprise-grade security.
AI-powered health recommendation system for certified practitioners that suggests personalized wellness plans based on patient medical history and current condition. Matches patient data against a comprehensive catalog of health/wellness products to craft tailored care packages, ensuring compliance, safety, and individualized treatment.
· 9 min read
Explore the integration of LLMs and vector databases to enhance data analysis. Learn how to use Claude 3.5 Sonnet, LangChain, and Pinecone to efficiently retrieve and analyze CSV data, providing natural language insights with ease.
· 6 min read
In this blog post you will learn how to deploy an app with HTTPs on an AWS EC2 instance using Ngrok. This cost-effective solution leverages Ngrok's secure tunneling to provide HTTPs without the need for a domain name or load balancer.
· 7 min read
In this blog post you will discover the transformative synergy between Amazon Bedrock FM and Pinecone Vector Database, reshaping the essence of Document Question Answering (QA). Uncover how these technologies empower chatbots to seamlessly navigate complex documents, offering a glimpse into the future of intelligent customer support.
Questions, collaborations or project ideas? Send me a message, I read every one.
