Surya Human

Introducing Claude Opus 4.7, our most capable Opus model yet. It handles long-running tasks with more rigor, follows instructions more precisely, and verifies its own outputs before reporting back. You can hand off your hardest work with less supervision.

Someone built a transparent Mario game that runs OVER IDE so can play while waiting for Copilot to write code.

Postman would soon be dead. It would become a legacy product used by humans while development. Since, now AI is coding and integrating most of the stuff, postman would be phased out slowly as AI don't need UI to visualize api. ⚰️ @getpostman

I just gave Claude Code a rooted Android phone… It autonomously reverse-engineered Subway Surfers, hooked the coin logic, bypassed the anti-cheat, and gave itself UNLIMITED coins in ONE session.

it's honestly embarrassing that Chrome accidentally became the best PDF viewer while Adobe spent decades making Acrobat slower and more bloated

Computer use is now in Claude Code. Claude can open your apps, click through your UI, and test what it built, right from the CLI. Now in research preview on Pro and Max plans.

@0xlelouch_ Is it not just simple Indexing on username?

Typed a Gmail username once and the UI instantly said: “Username already taken.” I asked an ex-Staff Google engineer the same problem (he was director of engineering in a startup i worked at), “You’re not doing an Elasticsearch query on every keypress, right?” He laughed. “No. That’d be a crime.” My classy approach: 1. Keep an in-memory trie of reserved usernames. 2. Update it async (delta pushes), not per keystroke. 3. UI checks locally in O(k) where k = username length. Numbers (why this is feasible): 1. Assume 2B usernames, avg length 10 chars. 2. Raw chars = 2B × 10 = 20B chars. 3. Even if you store 1 byte/char (not true in a trie, but baseline) that’s ~20GB just for characters. 4. A trie is about prefix sharing, so common prefixes collapse hard. Real memory is “nodes + edges”, not “strings”. 5. If we model ~1 node per char worst-case: ~20B nodes. - If a node is 8 bytes (tight packed arrays, bitsets, offset indices; no pointers), worst-case is 160GB. - With prefix sharing, you can easily cut multiples of that depending on distribution (gmail-like usernames are not random). 6. Shard by first 2 chars (36 possible: a-z, 0-9). 36² = 1296 shards. - Worst-case per shard: 160GB / 1296 ≈ 123MB. - Suddenly “instant check” fits in memory per front-end pod or edge POP. Yes, you can also do it with WebSockets: 1. Client streams “candidate username” events. 2. Server replies with availability. 3. Works fine, but now you’ve built a hot, stateful, low-latency service for… a UI hint. Most people will ship: 1. Elasticsearch prefix search. 2. Debounce 150ms. 3. Cache a bit. 4. Pray at peak signup traffic. And it works. But the trie approach is the kind of solution where the UI feels like magic tbh and it's something novel that i thought of. Things are just different at google scale.

Meet the new Stitch, your vibe design partner. Here are 5 major upgrades to help you create, iterate and collaborate: 🎨 AI-Native Canvas 🧠 Smarter Design Agent 🎙️ Voice ⚡️ Instant Prototypes 📐 Design Systems and DESIGN.md Rolling out now. Details and product walkthrough video in 🧵

@minchoi Any other competitors in ai ui design?

This Google Stitch Update is pretty wild. Insane app screens, landing pages, and UI flows from one prompt. 8 wild examples. Bookmark this.

mai iss chashmish ka gala daba dunga

--dangerously-skip-permissions

IT company CEOs to Claude:

I don’t get why Anthropic hasn’t used Claude Code to rewrite their bloated Electron app as native. Should be much easier than a C compiler.

We are literally about to build super intelligence before Microsoft ever figures out how to fix this

An arms race is about to begin

Our intern just asked me why we don't use Kubernetes. I said because we don't need Kubernetes. He said everyone uses Kubernetes. I said everyone TALKS about using Kubernetes. Most companies are running Docker containers on three servers and calling it a day. We have 40 employees. Our entire infrastructure runs on AWS with auto-scaling groups. It works fine. Kubernetes is designed for companies running thousands of services across hundreds of servers. We have twelve services. But he read that Kubernetes is "industry standard" so now he thinks we're behind. This is what happens when people learn from tech Twitter instead of actual experience. They think every company is Google-scale and needs Google-scale solutions. We don't need Kubernetes. We need our MySQL database to stop running out of connections because someone wrote a query that doesn't close properly. But that's not exciting. Nobody writes blog posts about "I fixed a connection leak." They write about "How we migrated to Kubernetes and saved millions" even though the migration cost more than they saved. I told the intern he should learn why tools exist before learning the tools themselves. He looked disappointed. He wanted to put Kubernetes on his resume.

@VasiliyZukanov Do you think the hype about claude is real? I've also been using the cursor for a while and it has fulfilled my requirement, in what comparison is claude better? I've tried simple prompts and both cursor 20$ plan and claude opus 4.5 gave me similar results. What's the hype about?

@Sppro20 I use Cursor and Codex. Cursor's Composer-1 is pretty good for quick, local changes (e.g. UI adjustments). For larger scope, Opus and GPT flagships are considerably better.

Claude Opus 4.6 and Chat GPT 5.3 released 30 minutes apart. Christmas came early this year 🎊 P.S. LLMs rivalry is hilarious

@matt_elevenlabs 11B

We just raised $500M at an 11B valuation 🎉 To celebrate, we’re giving away 1,000 free credits so you can test our platform. For the next 6 hours, comment “11B” below and we’ll DM you the credits (must follow) 👇

today’s one-sentence horror: sudo has been largely maintained by a single person for ~30+ years