Daniele Giunta

This talk explores how to build a local-first semantic image search system that runs entirely on your laptop. Using modern Python tools and local models - such as CLIP, FAISS, BLIP, BERTopic, FastAPI, and Streamlit - we turn a photo collection into a searchable library using text or images.

There is increasing talk about local language models in the context of LLMs, but finding practical notions of fine-tuning them is not easy. This talk will offer a concrete example for customizing a local LLM and adapting it to specific needs.