Google AI Edge Foresight: On-Device Meeting Notes with EmbeddingGemma 2
Melvin Vivas · X video post · 2026-10-21 · Open on X
Topics: Embeddings & Vector Databases, Retrieval-Augmented Generation (RAG), AI Dev Tools & Productivity · Level: beginner
Summary
Google AI Edge Foresight is an on-device macOS meeting companion. It uses the EmbeddingGemma 2 embedding model to improve shorthand meeting notes and search local documents, images, and transcripts in real time. It's an example of local semantic search over personal files that runs entirely on-device.
Key points
- Foresight runs locally on a Mac, so meeting notes and files stay on the device.
- It's powered by EmbeddingGemma 2, an on-device embedding model from Google.
- It automatically expands shorthand notes into fuller notes.
- It searches across local docs, images, and transcripts in real time while the meeting runs.
- The hard part it solves is connecting live notes to your existing local files, not taking the notes themselves.
Resources mentioned
- Google AI Edge Foresight · tool · developers.google.com · free
An on-device macOS meeting companion that improves notes and searches local files, images, and transcripts. - EmbeddingGemma 2 · tool · huggingface.co · free
Google's open 740M-parameter multimodal embedding model that puts text, images, video and audio into one shared vector space for on-device use.
Also in: Try EmbeddingGemma 2 multimodal embeddings in Colab (Melvin Vivas on X · notes), A day of work done from a phone with Grok Bot agents (Melvin Vivas on X · notes), EmbeddingGemma 2 Colab Notebook for FAQ Semantic Search (Melvin Vivas on X · notes), EmbeddingGemma 2 released on Hugging Face (Melvin Vivas on X · notes) and 2 more
Try this
- Build a local semantic search tool that embeds your notes, docs, and transcripts with an on-device embedding model such as EmbeddingGemma.