Designed and implemented a contract analysis chatbot using Retrieval-Augmented Generation (RAG) to help users review, interpret, and summarize legal documents with the assistance of AI. The system allows users to upload contracts and receive context-aware, evidence-backed answers to questions about clauses, risks, and obligations.
Built with a combination of OpenAI models, vector embeddings, and document retrieval pipelines, the chatbot intelligently references relevant contract sections to ensure accurate and transparent responses. It also provides summaries and compliance insights, significantly reducing the time and effort required for manual document review.
This project demonstrates how AI and NLP can streamline legal and enterprise workflows, making complex information more understandable and actionable for professionals and non-experts alike.

Designed and implemented a contract analysis chatbot using Retrieval-Augmented Generation (RAG) to help users review, interpret, and summarize legal documents with the assistance of AI. The system allows users to upload contracts and receive context-aware, evidence-backed answers to questions about clauses, risks, and obligations.
Built with a combination of OpenAI models, vector embeddings, and document retrieval pipelines, the chatbot intelligently references relevant contract sections to ensure accurate and transparent responses. It also provides summaries and compliance insights, significantly reducing the time and effort required for manual document review.
This project demonstrates how AI and NLP can streamline legal and enterprise workflows, making complex information more understandable and actionable for professionals and non-experts alike.

This project is an advanced AI-driven interview preparation platform designed to help users build confidence and improve their performance in any interview scenario. It enables candidates to practice interactively with an AI interviewer across three modes — chat, audio, and video — providing a realistic and adaptive simulation experience.
Powered by a combination of cutting-edge technologies, including OpenAI’s language models, Google Cloud, AssemblyAI, embeddings, and PostgreSQL, the system generates tailored interview questions based on the user’s target role or industry. During each session, it delivers real-time feedback and analysis on the user’s responses, assessing communication clarity, relevance, tone, and confidence.
Once the session is complete, the platform provides a comprehensive performance evaluation, summarizing key strengths and areas for improvement. A personal analytics dashboard visualizes overall progress over time, allowing users to track their growth, identify skill gaps, and prepare more strategically.
By combining natural language processing, speech recognition, and interactive UX design, this AI-powered coach offers an innovative, data-driven approach to professional development and interview readiness.

Contributed to the development of an innovative AI-driven social platform that enables users to create, share, and interact with personalized AI characters in friendly, conversational settings. Each AI character is designed with distinct personalities and behaviors, allowing users to engage in natural, emotionally intelligent dialogue experiences.
As part of the development team, I worked on integrating Stripe payment systems to enable creators to monetize their AI characters, providing a secure and seamless way for users to support their favorite creators. The project combines large language model (LLM) technologies, React-based front-end development, and Next.js services to deliver real-time conversational engagement in a dynamic, user-friendly environment.
This platform blends AI creativity, personalization, and digital economy innovation, fostering a new kind of interactive experience where technology meets storytelling and community.

Developed an AI-powered chatbot integration for Redash, a leading data visualization and analytics tool. This chatbot allows users to query and explore datasets using natural language, removing the need for complex SQL queries or manual filtering.

The system leverages LLM technology, Retrieval-Augmented Generation (RAG) principles, and Python-based backend integration to interpret user intent and translate queries into actionable data requests. By connecting with Redash’s API, the chatbot dynamically retrieves data, generates visualizations, and provides insights in real time.

This solution enhances data accessibility and user experience, empowering non-technical users to engage with data more intuitively. It represents a step toward conversational analytics, where AI bridges the gap between technical complexity and human understanding.