Rick Donato

Big news! 🤯

@charliermarsh @zeeg Oh I was so hopeful (for 30 secs) before I read @charliermarsh’s point around GCloud 😂

@zeeg I can’t believe you need a Google Cloud project to use this

Dang how did I miss this new Google workspace CLI (aka hello calendar) github.com/googleworkspac…

Big news 🚀🔥

✨🔥 System Prompts for Internal Tools & AI Models 👇

@AnubisEgx @taalas_inc I like the idea behind this tbf. However what’s the cost of the hardware?

@taalas_inc bro can’t solve simple math

24 dedicated people. $30M spent on development. Extreme specialization, speed, and power efficiency. Today we launch Taalas’ first product. Check it out: Details: taalas.com/the-path-to-ub… Demo chatbot: chatjimmy.ai API: taalas.com/api-request-fo…

Introducing Lyria 3, our new music generation model in Gemini that lets you turn any idea, photo, or video into a high-fidelity track with custom lyrics. From funny jingles to lo-fi beats, you can create custom 30-second soundtracks for any moment. See how it works. 🧵

✨The Free Social Platform for AI Prompts prompts.chat

Converts JSON Schema definitions into Pydantic models github.com/HideyoshiNakaz…

@svpino Well said 👍👍👍

I’ve been building software for 31 years, more than 20 professionally. I understand “architecture principles and best practices”. No, you can’t vibe-code Salesforce, SAP, or any other complex enterprise app.

Google just dropped "Attention is all you need (V2)" This paper could solve AI's biggest problem: Catastrophic forgetting. When AI models learn something new, they tend to forget what they previously learned. Humans don't work this way, and now Google Research has a solution. Nested Learning. This is a new machine learning paradigm that treats models as a system of interconnected optimization problems running at different speeds - just like how our brain processes information. Here's why this matters: LLMs don't learn from experiences; they remain limited to what they learned during training. They can't learn or improve over time without losing previous knowledge. Nested Learning changes this by viewing the model's architecture and training algorithm as the same thing - just different "levels" of optimization. The paper introduces Hope, a proof-of-concept architecture that demonstrates this approach: ↳ Hope outperforms modern recurrent models on language modeling tasks ↳ It handles long-context memory better than state-of-the-art models ↳ It achieves this through "continuum memory systems" that update at different frequencies This is similar to how our brain manages short-term and long-term memory simultaneously. We might finally be closing the gap between AI and the human brain's ability to continually learn. I've shared link to the paper in the next tweet!

Uptime monitoring dashboard for websites and APIs

Jeff Bezos heading to AWS headquarters tomorrow

Converts PDFs to structured text with table, handwriting, and multi-column support

ChatGPTs Codex CLI is good. Surprisingly good. 🤯

Remember, I'll be sharing my screen tomorrow to build some agents and help you understand how everything works. This is all free. You just need to join a Google Meet, and that's it. No sign-up necessary. You don't need to give me your email. Nothing. You'll find the Google Meet link in the post below.

@tomasrollo @willmcgugan More lightweight. I’d prob always go for Pydantic tbf so you can validate at runtime rather then when running mypy etc

@willmcgugan Any advantages over Pydantic classes? (Esp for the JSON API use case?)

Python's TypedDicts are awesome for taming the chaos that is JSON APIs.

@willmcgugan Why not use Pydantic?

@JulianGoldieSEO YES

Screw it. Leaking my entire AI playbook 🧠💸 You’re getting everything I use daily: ✅ 200+ elite prompts for GPT-4, Claude, Grok & more ✅ n8n automations for client outreach + hands-free SEO ✅ exact systems I used to scale ✅ 1,000+ high-converting copywriting angles, hooks & scripts Basically… if you’re serious about making money with AI — it’s all inside. LIKE + RETWEET + REPLY “YES” and I’ll send it over.