Developer Keynote recap – Richard Seroter
summary
Gemini 1.5 Pro Vision claims it can ingest a 75‑minute keynote and extract who spoke when and demo times, not just a transcript, and BigQuery Studio now runs continuous queries and multimodal SQL against images and text in one environment. Retrieval Augmented Generation chunks data and uses vector search across Google Cloud databases to boost trust in LLM outputs, and developer tooling like Gemini Code Assist and Gemini Cloud Assist aim to cut ten separate products down to a single simpler workflow.
AI-written summary · about
chapters (7)
- 0:00Intro
- 0:56Gemini Pro Vision
- 2:09How Gemini is changing software delivery
- 4:18Gemini in BigQuery
- 5:28What is Retrieval Augmented Generation (RAG)?
- 6:53Closing thoughts
- 8:20Wrap up
related talks
01h 16mI'm in my Gemini Era!
19mGemini 2.0: Beyond the Basics
18mLet's test Gemini, Google's AI in an Android application!
01h 01mBuilding Multimodal Apps with Gemini 2.0 and Gemini SDK
56mBuilding Multimodal Apps with Gemini 2 0 and Gemini SDK
01h 45mAI Engineering with the Google Gemini 2.5 Model Family
18mIntroduction to Gemini APIs and AI Studio
02h 58mDeveloper keynote deep dive