I am a Generative AI Specialist
Photo by rishi on Unsplash
Welcome to my portfolio! I am a Generative AI and NLP specialist with a strong foundation in reliability engineering and over 9 years of experience driving AI transformation across industrial and enterprise domains. My expertise lies in building intelligent systems powered by Large Language Models (LLMs) like GPT-4 and LLaMA to solve real-world problems—ranging from document understanding and knowledge retrieval to predictive maintenance and anomaly detection. I have designed and deployed advanced AI tools such as contextual chatbots, automated summarizers, and natural language interfaces that enable seamless interaction with structured and unstructured data. By leveraging Retrieval-Augmented Generation (RAG), prompt engineering, and semantic search, I’ve created systems that extract insights from complex data sources like engineering logs, maintenance manuals, and real-time sensor streams.
My journey began with ensuring the reliability of physical systems, and evolved into architecting robust AI-driven solutions that deliver business value. I thrive at the intersection of engineering and machine intelligence—where accuracy, interpretability, and innovation meet. Whether it's extracting intent from user queries or constructing dynamic API pipelines using NLP rules, I bring both depth and creativity to every solution. Through this portfolio, I invite you to explore how Generative AI can reshape industries, enhance decision-making, and unlock next-generation experiences. Let’s build the future—intelligent, reliable, and conversational.
A fun and intelligent emotional support bot powered by a Open Source LLM, capable of understanding user input and generating empathetic, mood-aware responses. A great showcase of applied NLP and emotional intelligence in AI!
A lightweight LLM-based RAG assistant built using opensource LLM and React for frontend that can answer domain-specific queries and demonstrate prompt-based interaction.
Built and deployed domain-specific conversational agents using GPT-4 and LLaMA to provide real-time technical support, document Q&A, and contextual assistance.
Crafted tailored prompts and fine-tuned LLMs to enhance output relevance across tasks like summarization, Q&A, and information retrieval.
Integrated LLMs with vector databases such as FAISS and Pinecone to build scalable systems for accurate and contextual information generation.
Designed systems to extract user intent, key entities, and rules from natural language queries to dynamically construct API calls and retrieve data.
Automated report generation and log summarization using LLMs to boost productivity and enable actionable insights from unstructured data.
Delivered full-stack GenAI applications with backend orchestration, LLM inference, and frontend UI integration using Streamlit, FastAPI, and cloud tools.
Click the below link to read full blog!
Click the below link to read full blog!
Click the below link to read full blog!
Click the below link to read full blog!
The photo credits for the tiles above are given in the respective blogs.