Vimerse.app is an innovative AI-powered video creation platform that helps users transform ideas into engaging, professional-quality videos with ease. It offers an all-in-one workflow where users can create characters, generate scripts, add realistic AI voiceovers, design scenes, and export complete videos in just a few steps. By integrating advanced AI technologies, Vimerse simplifies the video production process, making it ideal for marketing, education, storytelling, and social media content. The platform gives creators full creative control while saving time and effort through smart automation.
AI BookWriter is an advanced AI-driven platform developed using Python, OpenAI’s large language models (LLMs), LangChain, and Django. It automates the end-to-end process of book creation by allowing users to input a title, desired number of chapters, and target word count. The system then generates complete, contextually accurate books, often exceeding 100,000 words, within minutes. Designed to handle a wide range of content types, including storybooks, technical manuals, and digital magazines, AI BookWriter ensures high-quality output with speed and precision.
I developed an AI Clinical Visit Planner web app that helps doctors efficiently manage daily patient visits. Using LLMs (via DSPy) and Google Maps Distance Matrix API, it prioritizes urgent patients while minimizing travel time. Multiple AI agents analyze doctor availability, patient urgency, and locations to generate optimized routes. Doctors can input clinic hours and patient details to receive a clear, prioritized schedule with built-in navigation support.
Developed a domain-specific Q&A chatbot using Retrieval-Augmented Generation (RAG) by integrating OpenAI LLMs with a Vector Database (FAISS/Pinecone) and LangChain. The system enables accurate, real-time responses from custom documents such as legal contracts and company policies. Built with a modular, scalable architecture, it enhances knowledge retrieval and automates insight discovery for enterprise use cases.
Developed Zestya AI Assistant, a real-time conversational AI agent designed specifically for a fitness coaching application. The system enables natural, voice-based interactions with users, providing personalized workout recommendations and guidance based on their fitness history and goals. It integrates OpenAI’s GPT-4o Realtime API and leverages LangGraph to manage conversation flow and memory. The assistant offers context-aware, intelligent support to enhance user engagement and deliver real-time fitness coaching experiences.
