Samridhi

Hi @X algorithm. I'm Samridhi, here to #connect with people working on: ✔️Backend Systems & Data Engineering ✔️AI Tools & Applications ✔️Cloud & Distributed Systems ✔️Python / SQL / PySpark ✔️Developer Productivity ✔️Tech x Policy / Geopolitics ✔️System Design ✔️Interview questions If you're building, writing, or thinking deeply about tech, let's connect! If you're building in tech or want to exchange ideas, let's connect! #Tech #Al #SaaS #Networking #SoftwareEngineering #BackendDevelopment #CloudComputing

@claudeai Phew... This was the exact tweet I was waiting for the entire day.

As before, you can use up to half of your weekly usage limit on Fable 5. After that, you can continue using Fable 5 with usage credits, or switch to another model to keep working within your remaining limits. More details here: support.claude.com/en/articles/15…

We're extending Claude Fable 5 access on all paid plans, as well as keeping Claude Code’s weekly rate limits 50% higher, through July 19.

@GoogleCloudTech Great and cool new format to improve data sharing for AI systems 😎

In case you missed it: We recently introduced the Open Knowledge Format (OKF), an open specification that formalizes the LLM-wiki pattern into a portable, interoperable format. How OKF works → goo.gle/3R8Pry7

@elonmusk Great move. It's important to cater to the previous generation and onboard them before we think of catering to the current or next generation. Tech advancement should uplift all, instead of just a handful.

Tesla self-driving enables safer mobility for senior citizens

The most common failure in production AI apps isn't hallucination. It's HTTP 429. Datadog traced millions of LLM calls: ~5% error out and 60% of those errors are rate limits. Which means the hardest problems in AI engineering are ones backend devs already solved years ago: exponential backoff, jitter, request queues, semaphores, caching. The models are new. The engineering is not.

@sama @iliketeslas Apple sues and suddenly they’re “s-tier” and he’s posting like a fanboy. Totally not terrified though.

@iliketeslas i am not afraid of apple, but i have tremendous respect for them. s-tier company.

Sam Altman wasn't afraid of Elon but he is terrified of Apple. You can tell by all his posting today

@ClaudeDevs That's an amazing upgrade. I will try it soon.

Claude Code on desktop now has an in-app browser. Claude can pull up docs, designs, or any other site. It can read, click through, and interact the same way it does with your local dev servers. It's sandboxed and configurable: you choose whether sessions persist.

@buildanythingso Is this hackathon for testing the best AI model right now, or should I skip it?

hackathon in 2 days reply and we’ll send you a DM

@AravSrinivas We will run out of water trying to keep the data centers cool by then 😅 good luck to us!!

Imagine a fable 5 quality model that’s 3-4x less expensive in less than 6 months. And an Opus 4.8 grade model that can run on a local device in less than 12 months. Greater than 50% chance that these events will happen. Worth keeping in mind when you make predictions about the future.

By “this” he means scamming 🤣🤣

@elonmusk It's crazy seeing two adults- mature and highly successful men, being so mean to each other.

He takes scamming to a whole new level

@sama It's crazy seeing two adults- mature and highly successful men, being so mean to each other😉

there are a lot of benchmarks that suggest 5.6 sol is the best model in the world right now, but the most reliable way to tell is that elon is obsessed with me again

@sama It's actually worth it. I have been sticking to it like glue since its release.

makes us happy to see people love 5.6 sol so much!

@elonmusk It is commendable to observe the CEO of X actively participating in the marketing efforts for their product. This demonstrates a significant level of engagement and commitment.

Grok places second after Fable on real-world software engineering

@HarvardBiz Massive progress since 1965, no question. The persistent differences in experiences usually come down to voluntary choices and average group differences in competitiveness, work preferences, and family trade-offs, not mysterious double standards. Data keeps showing this.

Attitudes about women in leadership have improved since 1965. But there are still major differences in how men and women perceive evaluation, standards, and opportunity today. s.hbr.org/450eYNe

@elonmusk I'm building something for sure today!

Free trial of Grok 4.5 model via Grok Build CLI

@elonmusk Exactly — the scale is almost incomprehensible. Space solar changes everything

If solar power in space is harnessed for useful work at scale, the value of that will far exceed the Earth economy, because all of Earth civilization uses much less than a trillionth of our Sun’s power

@sama It's now my preferred model in Cursor AI as well!!

GPT-5.6 is now the preferred model in Microsoft 365 Copilot openai.com/index/gpt-5-6-…

@thsottiaux I'm impressed by the way every tech giant is extending usage to make sure everyone gets a taste of their model!

To celebrate the launch of GPT-5.6 Sol, we will reset the rate limits again (twice) across ChatGPT Work and Codex over the next 24 hours. We want you to have the time to truly try ambitious tasks and get the hang of it. Happy exploring!

@BratDotAI My claude and codex are self sufficient. They can work independently - write the code and tell me what they changed, why they changed, reason for error, etc. OH DEAR, I so love these AI babies 👶

Codex writes the code. Claude explains why it’s wrong. Perfect corporate team.

Same waiter. New kitchen — and this one runs on AI. 🧑‍🍳 AI calls are the slowest dish on the menu. A database query cooks in milliseconds — an LLM call takes seconds, sometimes tens of them. And right now, every table is ordering that dish. This is async I/O's dream workload: fire off the model call, go serve ten other tables while the tokens cook, come back the moment they land. Ten calls that would take 30s back-to-back finish in ~3. Embeddings, retrieval, tool calls, agents fanning out in parallel — modern AI apps are almost pure waiting. Which is exactly where async turns idle time into flow. The more tables stuck waiting on the kitchen, the less time anyone stands still. That's async in the AI era. There's always another table waiting on the kitchen. The floor never goes quiet. ✨