Chennai, Tamil Nadu, India
Summary
Contributed to advanced AI research and development at IIT Madras, focusing on multimodal chatbot systems and data pipeline engineering.
Highlights
Engineered and implemented a cutting-edge multimodal chatbot, capable of processing and interpreting diverse user queries across text, images, videos, and PDF documents.
Integrated advanced transformer-based models, including Qwen2.5 Omni and Sentence-Transformers, to enable context-aware text generation and embedding extraction for multi-turn conversations.
Developed robust custom pipelines for efficient image, video frame, and PDF text extraction, ensuring seamless multimodal input processing and highly accurate retrieval-based response generation.
Utilized a comprehensive tech stack including Python, PyTorch, Hugging Face Transformers, Gradio, OpenCV, PyMuPDF, FAISS, and NumPy for system development and optimization.