Fusion of multimodal data and insights generation using LLM and VLM Documentation
May 2023 - Apr 2024 (1 year)
• Created a pipeline to handle real-time multimodal data with Apache Airflow with optimised search to get big data in 5 mintues • Fine-tuned vision language model for deep fusion of multimodal data. Wrote prompts for llama model to gather insights from data. • This system helped to get meaningful insights from multi modal data with 26% accuracy