This is in essence how any "AI agent" in 2025 works

I cracked the code on how to generate the best Veo 3 video with Gemini API. Here are 10 video generate from a single short prompt, automatically improved with Gemini 2.5! No cherry-picks! 🤯🤯 The Secret? A Meta JSON Schema and let Gemini 2.5 generates high quality prompts with

This week on Construction Physics: where are vacation homes in the US? construction-physics.com/p/where-are-va…

Here is how to do multimodal (PDF/Images) few-shot prompting with Gemini 2.5: ❌ Don't add your few-shots in the last "user" message ✅ Send alternating (multi-turn) messages using "user" and "model" turns. Let's say you have 2 PDFs and 2 example outputs of what you want the

The beatings (free books) will continue until everyone looks at their data: 1. LLM Evals FAQ: maven.com/p/412512/frequ… 2. Beyond Naive RAG: Practical Advanced Methods maven.com/p/945082/beyon…

8 interesting ways people are using nano-banana (24 hrs in) - what else?

At Microsoft we are hosting a 9 part series covering the fundamentals of Python + AI! You'll learn about LLMs, RAG, MCP, Agents, Vector Embeddings and a lot more! Each class will have demos, code, slides and live instruction plus a weekly space for QA in our office hours.

How to build with Nano Banana Pro: Complete Developer Tutorial by Patrick Loeber. 🍌🍌 This guide will cover: 1. Using Nano Banana in AI Studio 2. Project setup 3. Image creation 4. Image editing 5. Photo restoration 6. Multiple input images 7. Conversational image editing 8. Best

Did you know SQLite has a vector extension? 🧮 SQLite is the most used database in the world and runs on almost any device. You can now easily build AI applications leveraging SQLite-vec and the new Embedding Gemma directly on-device, no internet required. Below is an simple

I’ve been sharing NotebookLM tips here for a while. Just compiled them all into one article so it’s easier to follow. All these ideas come from my personal experiments.

my biggest codex power user tip is to use dictation install macwhisper or whatever you like you'll just end up specifying like 3× more than you would by typing and avoid so many footguns just try it

I just learned Claude's new code interpreter mode has a /mnt/skills/public/ folder full of prompt instructions and Python utilities for creating and manipulating pdf, docx, pptx, and xlsx files - and you can ask Claude for a copy and learn a TON about working with those formats

New massive update Google TPUs Inference for open Models, like Gemma! `tpu-inference` isa new vLLM backend that delivers up to 5x performance gains over previous prototypes while supporting both frameworks via a single lowering path. - 🤝 Unifies PyTorch and JAX under

Exciting expansion! Waymo now serves the whole SF Bay Area Peninsula from SF to San Jose and is taking riders on freeways. waymo.com/blog/2025/11/t…

This is great, I'd love to see the same feature from OpenAI Codex Cloud and Claude code for web

This is wild. Nano Banana Pro just completely changed the AI image game. These are not real 🤯 6 wild examples (prompts included): 1. Polaroid style photographs

Excited to share a System Instructions for Gemini 3 Pro that improved performance on several agentic benchmarks by around 5%. 🚀 We collaborated with the Google DeepMind post-training research team to include some best practices in our docs. 🤝

Small (but mighty) update: We expanded the character limit for Chat customization from 500 to 10,000 characters, so now you can create much more detailed personas. Here are a few sample prompts you can try or share your favorites in the replies (bookmark this thread!) 1. The

🙌 Gemini 3.0 Flash is finally here! Flash is superior to Gemini 2.5 Pro in speed and performance (I've found it to be even better than Gemini 3.0 Pro, in some use cases! 🤯) - and it's super, super speedy and cost-effective. 👇Try it in Google AI Studio Build today!

The true size of the United States.