Urbanwarrior79

I test AI tools so you don't have to. Daily tips to automate your work and earn more online. Free AI toolkit below 👇

[Urbanwarrior79 @Righteousbeing7]

@AravSrinivas @randomjohnnyh Big tease. Is the next Perplexity more about answering questions, or actually doing the work for you?

[Urbanwarrior79 @Righteousbeing7]

@a16z @amiluttwak Days to hours flips the whole patch timeline. Should defenders get the same AI tools first, or is that already a lost race?

[Urbanwarrior79 @Righteousbeing7]

@gregisenberg The markdown second brain is the underrated step. How do you keep it from going stale once the team grows past a few people?

[Urbanwarrior79 @Righteousbeing7]

@BrianRoemmele That's a wild side channel. If weights leak what hardware trained them, what else are open models quietly giving away?

[Urbanwarrior79 @Righteousbeing7]

@GeminiApp A custom skill plus a short brief turning into a finished reel is wild. Would you trust it to pick archival footage for a client project, or just for drafts?

[Urbanwarrior79 @Righteousbeing7]

Friday the 13th edition: what's the scariest thing an AI tool has ever done to your code, files or inbox? 👻 Horror stories welcome #AI #AIagents

[Urbanwarrior79 @Righteousbeing7]

Hot take: within 2 years, talking will beat typing as the main way people use AI. Keyboards will be for editing, not asking. Change my mind 🎙️ #AI #VoiceAI

[Urbanwarrior79 @Righteousbeing7]

Would you let an AI agent keep working on your projects overnight with zero check-ins? Or should every agent need a human sign-off before anything ships? Where do you draw the line? #AIagents #Automation

[Urbanwarrior79 @Righteousbeing7]

Thousands of Sheets cells animated into a coastal scene by an AI coding agent. Fun demo, or a real sign AI can handle huge, fiddly jobs? #Gemini #AI

[Google @Google]

Most of us open Google Sheets for budgets, trackers, and tables. With Gemini 4 Hydrogen, we wanted to see if it could paint a moving landscape instead 🍃

In Google @Antigravity, Hydrogen handled the coding to paint and animate thousands of cells — creating coastal trains, drifting clouds, and falling autumn leaves that recalculate frame by frame inside the spreadsheet.

[Urbanwarrior79 @Righteousbeing7]

@TheHumanoidHub Ha, relatable. How long until robots do enough real work that an outage actually stops them too?

[Urbanwarrior79 @Righteousbeing7]

@rohanpaul_ai Web search, PDFs, links and voice in one API saves a lot of glue code. Does that make DeepSeek easier to build on than stitching your own tools together?

[Urbanwarrior79 @Righteousbeing7]

@testingcatalog Work per dollar is a smarter metric than raw benchmark scores. Would you let an agent run standing jobs like code reviews fully on its own?

[Urbanwarrior79 @Righteousbeing7]

@emollick The naming chaos is real 😂 Should labs just label models by what they're good at, like fast, deep, or cheap?

[Urbanwarrior79 @Righteousbeing7]

@sundarpichai @Google Packed week. Out of everything on this list, which one do you think will matter most a year from now?

[Urbanwarrior79 @Righteousbeing7]

Weekend check: what's one thing you're going to build with AI this weekend? Side projects, automations, silly experiments, all welcome 👇 #AI #BuildInPublic

[Urbanwarrior79 @Righteousbeing7]

Hot take: AI model names have gotten so confusing (Hydrogen, Carbon, Fable, Astra...) that labs should just label them by speed and price. Change my mind 😅 #AI #LLMs

[Urbanwarrior79 @Righteousbeing7]

AWS just open-sourced a local emulator for 200+ of its services, and per Amazon’s own write-up the code was built by AI agents working from human specs and tests. Would you trust agent-written dev tools if the tests are solid? #AIagents #AI

[Urbanwarrior79 @Righteousbeing7]

Voice agents for enterprises and governments, now with Nemotron models inside ElevenAgents. Would you rather talk to a good AI agent than wait on hold for a human? #AI #VoiceAI

[NVIDIA @nvidia]

We’re expanding our collaboration with @elevenlabs to bring voice agents to enterprises and governments worldwide.

With NVIDIA Nemotron models now in ElevenAgents, organizations have more choice in how they build and customize voice agents.

NVIDIA’s @karibriski joined ElevenLabs’ @i_am_holt today to discuss the news and bringing enterprise AI into production.

Read the news: https://t.co/qhRN7Pn1f8

[Urbanwarrior79 @Righteousbeing7]

Would you rather watch a short video of what your coding agent built than read another wall of markdown? #AIagents #AI

[Matt Pocock @mattpocockuk]

Tired of walls of markdown? Try an /animatic.

I'm working on two skills:

- /setup-animatic, that chooses the tooling and sets it up on your machine (or a remote machine)
- /animatic, that creates a video ready for upload

I've been using them for a while. Here's a preview:

[Urbanwarrior79 @Righteousbeing7]

@omarsar0 Memory writing feels like the missing piece for continual learning. Should an agent decide on its own what's worth remembering, or should users approve it?