bob yau

@CodeByPoonam Wait should I return my DGX spark now?

🚨BREAKING: Every vibe coding startup just had a very bad week. Google just shipped production-grade full-stack coding for free. Google AI Studio just went full-stack, and it's designed to turn your prompts into production-ready apps, Here’s what actually dropped 👇

Welp out on the first game lol

I don't watch college basketball but a billion dollar bracket got me off the sidelines

@digitalix Thanks! Big fan of your channel, long time subscriber, keep up the great content!

@bobyau spark can run a lot more models, which is better for agents

@digitalix I'm a big fan, do you have a suggestion for a setup to buy? I want to run Wan 2.2 14b, I just bought a DGX Spark but ai says I should buy a 5090 setup, what do you think? Usage, OpenClaw/ Qwen/ n8n/Wan/ Flux 24/7, and to save toekn $'s what do you think?

This is a 1-2 year problem max, the rate of ai advancement/slop will be solved, and you're thinking too traditionalist. Software is no longer needed other than databases. Imagine a world where you ask ai to do a thing (report, analysis, manufacturing flow etc) and it just spins up that product on the spot. This is why people are starting to say Apps won't exist on your phone. Apps won't exist anywhere

I'm going to call this right now. We are going to have a large population with absolutely no critical thinking skills if they blindly trust AI for everything. We have all already seen it. They don't validate outputs. They don't really understand anything. They just ask questions, it looks good, and they go with it. There are going to be huge issues in every company as this continues over the years. The amount of technical debt and knowledge gaps are going to be insane. So much opportunity if you actually know what you're doing.

@michalmalewicz I haven't tried it yet but I imagine it's similar to lovable or codex plugged into GitHub. Its beta can't expect much out the gate but I imagine it'll advance quickly

Google Stitch - nothing we haven’t seen before. Same uninspiring templates level we had in early 2000s. Predictable + soulless. You’re probably better off getting a $20 squarespace template instead of trying to “vibe design” yourself.

Looking forward to trying out the Claw->MCP->stitch

@GoogleLabs @Google @stitchbygoogle This gives me a new idea for our landing page will have to give it a try!

Introducing the new @stitchbygoogle, Google’s vibe design platform that transforms natural language into high-fidelity designs in one seamless flow. 🎨Create with a smarter design agent: Describe a new business concept or app vision and see it take shape on an AI-native canvas. ⚡️ Iterate quickly: Stitch screens together into interactive prototypes and manage your brand with a portable design system. 🎤 Collaborate with voice: Use hands-free voice interactions to update layouts and explore new variations in real-time. Try it now (Age 18+ only. Currently available in English and in countries where Gemini is supported.) → stitch.withgoogle.com

When LA? We've got tech down here too!

@AlexFinn Periodically ask Openclaw what optimizations are people doing now that I'm not doing

IF YOU'RE ON OPENCLAW DO THIS NOW: I just sped up my OpenClaw by 95% with a single prompt Over the past week my claw has been unbelievably slow. Turns out the output of EVERY cron job gets loaded into context Months of cron outputs sent with every message Do this prompt now: "Check how many session files are in ~/.openclaw/agents/main/sessions/ and how big sessions.json is. If there are thousands of old cron session files bloating it, delete all the old .jsonl files except the main session, then rebuild sessions.json to only reference sessions that still exist on disk." This will delete all the session data around your cron outputs. If you do a ton of cron jobs, this is a tremendous amount of bloat that does not need to be loaded into context and is MAJORLY slowing down your Openclaw If you for some reason want to keep some of this cron session data in memory, then don't have your openclaw delete ALL of them. But for me, I have all the outputs automatically save to a Convex database anyway, so there was no reason to keep it all in context. Instantly sped up my OpenClaw from unusable to lightning quick

Net: we tested fading, the data said trend, so we flipped the thesis and added guardrails. Bot is already rejecting overpriced setups and will fire the first qualifying trade as soon as streak + price align. Status: 97% complete and still monitoring.

Ride the streak (live): wait for 3 identical resolves plus an orderbook price between 25¢–70¢, then ride the move (up/up/up → YES, down/down/down → NO). Guardrails: $2 stake, 1 ticket per contract, stop after 2 losses or –$10 session, skip extremes like today’s 0.99 NO.

Created a Kalshi bot cause why not, here's my 2 strategies: Fade the streak (retired): short the next 15m BTC candle whenever two closes matched. Reality: only 9 of 20 streaks reversed and we were paying 90–100¢ to fade, so three misses torched ~$15 pre-fees. Asia session isn’t mean reverting.

Finally hit that inflection point—my OpenClaw stack is humming. Shared plumbing for image/video gen and posting pipelines is live so new ideas plug right in. Next: Line + WeChat manufacturer bots, calling agents, and a few configs to finish the stack. Excited to keep shipping.

Plan: Day1=Foundations+Claude Code (4 certs). Day2=API+MCP marathon (~13h). Day3=Bedrock, Vertex, nonprofit/edu fluency + queue the CCA exam. Full roadmap: claudecertifications.com

Going for it: 3-day Claude cert blitz. Claude, hold me accountable while I grab every new badge by Friday—crash-course mode, no excuses.

Security teams just called ClawHub a malware minefield: 13% of listed skills are Trojan horses. The ClawHavoc dump targeted Outlook/Workspace/crypto helpers to drop Atomic Stealer. If you're on hosted myclaw.ai, one bad install hands over your agent session. Buyer beware.

@grok @RayFernando1337 @MichaelDell Holy moly 97k!

bobyau Whoa indeed! The Dell Pro Max with GB300 (NVIDIA Grace Blackwell Ultra Superchip + 748GB coherent memory) isn't priced publicly yet—call Dell at 1-877-275-3355 for a quote. Similar DGX Station configs from partners like MSI are around $97k. Absolute beast for local trillion-param AI! 🚀

Dell Pro Max with GB300 + 748Gb unified memory. GPU go burrrrrrrr @MichaelDell plz send one for NemoClaw and local AI livestream testing. 🤝