PixelShipper

@jayvraavi Kudos, that’s awesome. How long did that journey take?

100K reviews on App Store 🥳

@moonfarm_dev In the same no sleep boat. People without youngins have no idea what it means to actually be tired. Godspeed

Good morning! Working on the final touches of the new landing page and complete rebranding of the entire Mara platform 🥳 also battling poor sleep because kids have fever and don't sleep well 😅 What are you doing today?

@Gamingtronium If only I got a dollar for each time I saw this post

Claude Opus 5 is insane. i know literally NOTHING about coding. ZERO. and i just built 3 fully functioning web apps in 30 minutes. http://localhost:3000/ http://localhost:8000/ http://localhost:5000/ check it out.

@ChadAppDev Nice bump though

@PixelShipper Nah old

I love ASO

@_catwu I think the tokens per second are an issue at the moment

Claude Opus 5 is great at long-running autonomous work! Try it out and let us know what you think :)

@ChadAppDev How many back and forths did you have with them?

My infamous app has been accepted after 1 month

@moonfarm_dev Can confirm it works well (so far). Only tested within one project repo and compared to my existing wiki structure. Definitely more elaborate than my more simplistic structure.

@PixelShipper Cool! I might try it! Did you?

Nobody writes docs anymore. The fix, apparently: let the agents write their own. OpenWiki (12.8k stars in its first month) is a CLI from the LangChain team that generates and maintains a wiki for your codebase, written for agents rather than humans. One command creates it, and a CI workflow opens a PR whenever the docs drift from the code. The sleeper feature is personal mode: it builds a local "brain" wiki from your own sources: Gmail, Notion, web search, Hacker News, X — that any of your agents can read. npm i -g openwiki, then openwiki --init. That's the whole setup. Docs used to rot because humans wouldn't maintain them. Agents don't get bored. Have you tried it yet?

@bcherny Overloaded at the moment? Super slow right now

Opus 5 is a great model for coding, data analysis, design, biology, knowledge work. More than any of these eval scores, what is most exciting to me is something else: Opus 5 is our least prompt injectable model yet. It is a bit buried in the system card, but across PI evals and red teaming, Opus 5 is very hard to prompt inject successfully. And when layering defenses -- strong model alignment, combined with prompt injection probes, combined with Auto Mode in Claude Code -- the success rate for prompt injection attacks drops to ~0. This is new and exciting! More about this soon. #page=73" target="_blank" rel="nofollow noopener">www-cdn.anthropic.com/c5fbac3f0b1280…

@elonmusk Yes he is, and good to see more industry leaders validating the importance of this

This has my full support. Jensen is right.

@LukaTheFounder Similar to one of my projects. Love the design and good job on the app store screens!

First internet money. Craziest feeling there is!

Great to hear about the importance of open models from one of the biggest industry leaders.

@sflorimm It’s mostly hype with rarely any meaningful return

still can’t figure out why people are launching on product hunt

Repo: github.com/langchain-ai/o… Two notes: scheduled/CI runs send anonymous reliability telemetry (one env var disables it), and the wiki output follows Google's Open Knowledge Format, so it's portable between tools.

@moonfarm_dev There is no coffee reference

My old profile vs my new and updated one anything you'd change? 🤔

@moonfarm_dev You sure have a knack for ui/ux on web. In my experience a demo is something that will help you out, and at the very least improve your lead funnel

@PixelShipper The one in the quoted post is the old one 😅 Hm, I should make a demo, good point! I'll launch the rebrand and then try to make a demo next week! 🙌

@ClaudeDevs Great, but where’s Opus 5?

On the Chartography benchmark (100 questions over dense real-world charts), Fable 5's accuracy goes from 29% to 73% with a zoom tool, and Sonnet 5’s from 13% to 44%. See the cookbook to learn more: github.com/anthropics/cla…

@DanielSmidstrup It will do that initially. Once it determines you are running it regularly in clamshell or simply plugged in, it will reduce charging to 80% to save battery long term life. Let it do its thing, it will handle the battery management perfectly fine.

@PixelShipper But it goes to 100% when plugged in. Is that just visually done now?

Genuine question. I got my first MacBook a few days ago. Do you leave yours plugged in all day? Or do you unplug it once it hits 100%?