Jon

@SouthDallasFood Molcajete! My go to when we have a night out for some Mexican.

Ok serious question, how does one even go about eating this?

@SlimFold @SlimFoldAS

6 years later and still rockin' the slim fold wallet

> this would be awesome to have in my batting cage

> be roblox > launch a gaming platform > 70M daily active kids click their screens to hatch virtual pets > every click is actually a proof-of-work hash calculation > harness planetary-scale compute to brute-force crypto wallets > pay the labor with digital hats

@ChShersh pull down to refresh

Imagine everyone in the world has your app installed on their phone. How would you implement a system to deliver 8B+ notifications to everyone within 2 seconds?

@suni_code keep it simple 😎

@jonathanmaestas Hahaha 😅

Interviewer: If a user uploads a 5GB file but loses internet halfway, how should the backend handle resuming the upload?

@SportsCenter @PeteThamel just change the background image to wood. no one would know.

The Big 12 is getting rid of the LED glass court 👀 (via @PeteThamel)

@jordymaui very creative approach! thank you for sharing!

@jonathanmaestas validation. the agent does its thing, then the script checks the output before it goes anywhere. stuff like "does this have the right format" or "did it actually include all 5 sections i asked for." if the check fails the agent has to redo it. examples coming in a thread.

your OpenClaw agent is ignoring half your instructions and you probably don't know it. Anthropic's new skills guide confirms what i learned the hard way - language instructions drift. you tell your agent "always validate the output" and it skips it 40% of the time. the fix is stupidly simple. write a script instead. a 10-line python file that checks the output beats a paragraph of instructions every single time. code is deterministic. language isn't. i moved all my critical checks from instructions to scripts inside the skill folder. overnight difference. the agent stopped cutting corners because it couldn't - the script either passes or it doesn't. if your agent is producing inconsistent results, this is probably why. stop asking it nicely. make it mandatory. i can drop a thread on this to teach you how to do this!

@karpathy @leerob @grok - what are your thoughts on this?

@karpathy @leerob should there be a stage between Tab->Agent? that seems like a big step or maybe not ? 🤔 what are your thoughts?

Cool chart showing the ratio of Tab complete requests to Agent requests in Cursor. With improving capability, every point in time has an optimal setup that keeps changing and evolving and the community average tracks the point. None -> Tab -> Agent -> Parallel agents -> Agent Teams (?) -> ??? If you're too conservative, you're leaving leverage on the table. If you're too aggressive, you're net creating more chaos than doing useful work. The art of the process is spending 80% of the time getting work done in the setup you're comfortable with and that actually works, and 20% exploration of what might be the next step up even if it doesn't work yet.

Google just dropped an official CLI for gmail, drive, calendar sheets and more Adding it to my openclaw right now

🔴 where do you get a ball like this? target?

@AlexChhk @marshallrichrds got it going. he's the one who got me to buy one of these phones to tinker with. cheaper than a pi and nearly as good.

@jonathanmaestas oh nice and can it do things like access other app data (e.g your photos etc or even like your browsing history and stuff) this way?

🦞 openclaw running on my phone

🔥 claude code running on my phone

@AlexChhk that's what i'm going to find out!

@AlexChhk nope. unrooted. using termux app, with its proot ability i'll keep you posted how it goes!

@jonathanmaestas did you have to root your phone