ALT Music

🚨 NEW DROP ALERT 🚨 That direct deposit just hit and we’re about to be a problem. 💸 "Paycheck Famous" drops TUESDAY 2/3. For everyone who’s a millionaire until Monday morning. 🍻 Pre-save link in bio. #NewMusic #CountryTrap #JaxsonGrit #ALTMusic #PaycheckFamous

@NeverteIImeodd Only security breach here was your AI made the whole door come off the hinges and slide all the way across the room...

@GraceGym_ No, he prompted an AI for hours to make a video.

He Built a Hidden Container House Under His Lawn 🚜

@akafaceUS I probably would have been the one landing on my face trying to use that ramp.

Anyone who jumped from the plane with their luggage in this situation should be added to the no fly list.

@DorLinder All these AI video fallout shelters are gonna make a lot of DIYers get themselves hurt.

Building a nuclear bunker in the backyard. But I really like the deck.

@googlemaps Childhood home...2007...my childhood is in the "before" times.

Time Travel 101: 1. Go to your childhood home on Street View. 2. Click “see more dates” 3. Slide the bar back to 2007. No need to hit 88mph.

@yanaqueen_ Seems the purpose of that action was to get attention.

Does anyone know the purpose of this action??

@LyalinDotCom What did you name it, if I may ask? Congratulations! If you haven't already, have your agent log everything! It helps my diagnostics of new skills much faster.

I created my own virtual assistant in about an two hours last night. It has persistent memory (SQLite), can read my email (0365 APIs I never used before), search the internet, and analyze web pages. It also talks to me. God, this is so cool.

@antigravity I couldn't agree more. With the capabilities of these agents, it's better for you to be the architect, let them be the builder.

Don't tell the Agent How. Tell the Agent What.

@lexfridman This is exactly why I built my agent myself. Security first, integration second.

The power of AI agents comes from: 1. intelligence of the underlying model 2. how much access you give it to all your data 3. how much freedom & power you give it to act on your behalf I think for 2 & 3, security is the biggest problem. And very soon, if not already, security will become THE bottleneck for effectiveness and usefulness of AI agents as a whole (1-3), since intelligence is still rapidly scaling and is no-longer an obvious bottleneck for many use-cases. The more data & control you give to the AI agent: (A) the more it can help you AND (B) the more it can hurt you. A lot of tech-savvy folks are in yolo mode right now and optimizing for the former (A - usefulness) over the the latter (B - pain of cyber attacks, leaked data, etc). I think solving the AI agent security problem is the big blocker for broad adoption. And of course, this is a specific near-term instance of the broader AI safety problem. All that said, this is a super exciting time to be alive for developers. I constantly have agent loops running on programming & non-programming tasks. I'm actively using Claude Code, Codex, Cursor, and very carefully experimenting with OpenClaw. The only down-side is lack of sleep, and an anxious feeling that everyone feels of always being behind of latest state-of-the-art. But other than that, I'm walking around with a big smile on my face, loving life 🔥❤️ PS: By the way, if your intuition about any of the above is different, please lay out your thoughts on it. And if there are cool projects/approaches I should check out, let me know. I'm in full explore/experiment mode.

@iyoushetwt I've been using Antigravity for everything lately. If you love VS Code, you'll love this too.

As a developer, which one do you prefer the most?

@GeminiApp This is kind of exciting! Since these songs will be watermarked, they won't be available for commercial use. But this will give me a testing zone for my music creation.

Here are some useful prompting tips to get the most out of our new music generation model in Gemini, Lyria 3 ↓

Appreciate it, Dmitry! 🚀 Short version: Our students saw a UNC Health report showing transit and healthcare access are huge barriers here. We're building Raspberry Pi community kiosks to solve it, and I custom-built a multi-node Firebase backend to test Gemini's integration as the brain. Shooting you a DM with the architectural breakdown!

@DavidTyson83 oh thats awesome id love to hear it, here or DMs are open

So we’re back… x was down right?

@ideklmao1_ @googlechrome They inadvertently killed it with the recent roll outs to prevent bots from using websites, it blocked that feature also.

@DavidTyson83 @googlechrome LMAO im so irritated it doesn’t work for me🫩

With auto browse in Gemini in Chrome (now available for Google AI Pro and Ultra subscribers in the US), you can delegate routine tasks like shopping, scheduling, and digital paperwork. It’s designed to keep you in the loop and ask for your sign-off, so you can focus elsewhere while it gets things done.

AI can diagnose a fault in milliseconds, but it can't climb a ladder, wire a PLC, or swap a bad sensor. In the trades, AI isn't a replacement, it's a force multiplier. It turns a 2-hour troubleshooting job into a 5-minute fix, but you still need human hands to finish it, for now.

Definitely interested. We built our own architecture ('Gemmy Core System') specifically because we saw a raw OpenClaw instance disable a friend's network adapter to 'optimize' a process. 💀 Does Pickle Rick have improved sandboxing to prevent that kind of self-sabotage? Would love to test the local LLM handoff.

If you folks have not checked out @galdawave Pickle Rick @geminicli extension yet, you're missing out something pretty damn amazing. He uses this to work on Gemini CLI, its a hidden superpower. We have plans for this too, but for now go check it out! github.com/galz10/pickle-…

@AustenMakers @antigravity Spot on, Mark. We’re wrapping all critical state updates in transactions for that exact reason. One race condition in the memory logs and the whole 'persona' glitches out. Appreciate the heads-up! 👊

@DavidTyson83 @antigravity That non-blocking flow is the holy grail for voice agents. With Firebase, keep an eye on write contentions if multiple agents hit the same doc simultaneously. Firestore transactions or atomic increments are lifesavers when concurrency spikes. Good luck with the stress test.

In Antigravity, you're not limited to one cursor. Spawn parallel agents to handle backend and frontend simultaneously and orchestrate like a pro.

This is the ultimate accessibility tool. We're testing Deep Think's ability to turn a napkin sketch not just into a 3D print, but into a functional solution for non-technical users. We want to bridge the gap where anyone—regardless of training—can solve physical problems just by visualizing them. That is how you upgrade quality of life. Logic is the tool, humanity is the goal.

Today, we’re releasing a significant upgrade to our specialized reasoning mode, Gemini 3 Deep Think. Deep Think is built to drive practical applications, enabling researchers to interpret complex data and engineers to model physical systems through code. With the updated Deep Think, you can turn a sketch into a 3D-printable reality. Deep Think analyzes the drawing, builds the complex shape, and generates a file so you can create the physical object with 3D printing. This is rolling out now to Google AI Ultra subscribers. Select the "Deep Think" option in the tools menu to get started. Learn more here: goo.gle/3MoiifF

This parallel architecture is exactly what we need for industrial safety. 🏗️ At ALT, we're building a 'Deep Think' agent that analyzes PLC wiring diagrams in the background while the frontend agent watches the student's hands in real-time via XR. If the background agent calculates a short circuit, it interrupts the voice feed to say: 'Don't flip that switch!' ⚡ Real-time safety needs deep reasoning.

We’ve upgraded our specialized reasoning mode Gemini 3 Deep Think to help solve modern science, research, and engineering challenges – pushing the frontier of intelligence. 🧠 Watch how the Wang Lab at Duke University is using it to design new semiconductor materials. 🧵