Abdul Baquee

@TheRedMike @Saurabh_Unmute Gore gore mukhde pe kala kala chashma, 😎 I love both Saurabh reporting. 😀

देखिए ईरान में Ali Khamenei के घर से @Saurabh_Unmute की Exclusive Ground Report, The Red Mike के यूट्यूब चैनल पर। #iran #Tehran

ANDREJ KARPATHY–INSPIRED FILE JUST HIT #1 ON GITHUB 82,100 stars. 7.8k forks. Zero dependencies. Not a framework. Not a tool. Just one CLAUDE.md file. And it’s changing how people write code with AI. Bookmark this. The insight comes from Andrej Karpathy LLMs don’t fail because they lack intelligence. They fail because they assume instead of clarifying, overcomplicate instead of simplifying, modify everything instead of focusing, and execute without clear success criteria. This file fixes that. Four principles Think Before Coding No assumptions. No silent confusion. State assumptions. Ask when unclear. Show tradeoffs. Push back when needed. It prevents wrong decisions before they happen. Simplicity First Minimum code that solves the problem. No overengineering. No unnecessary abstractions. No speculative features. If 200 lines can be 50, make it 50. Surgical Changes Touch only what is required. No drive-by refactoring. No random edits. Every changed line should map directly to the request. Goal-Driven Execution Define success before starting. Instead of “fix this,” write a failing test and make it pass. Turn tasks into verifiable outcomes and let the model iterate. Why this works Karpathy’s key idea is that LLMs are extremely good at looping toward a defined goal. If you give clear targets and verification, they improve themselves without constant guidance. What changes instantly Cleaner diffs with only relevant changes. Fewer rewrites caused by overcomplication. More clarifying questions before implementation. Simpler, production-ready code from the first attempt. Install Plugin /plugin marketplace add forrestchang/andrej-karpathy-skills /plugin install andrej-karpathy-skills@karpathy-skills Or download the file curl -o CLAUDE.md raw.githubusercontent.com/forrestchang/a… What this really does Claude becomes better at reasoning, which leads to better code, better systems, and real leverage. One line summary You don’t need a better AI. You need better constraints.

Stop telling Claude: “build this” Stop telling Claude: “write code” Stop telling Claude: “fix this bug” You’re using a staff-level AI like a junior intern. Claude performs best when you give: • role • constraints • architecture expectations • output format • real-world context Here are 10 production-grade prompts you can copy-paste 🧵👇

@IRANinMumbai When you rank 4th globally in IQ, the banter naturally hits different. 🧠 With the US sitting at 18th and India at 50th, I guess it takes a top-5 intellect to truly appreciate the difference between diplomacy and 'random bakwaas'. Levels to this game! 😏📉

Maybe someone should book a one-way cultural detox for Mr. #Trump, it might just reduce the random bakwaas 😏 Kabhi #India aa ke dekho, phir bolna.

Zinc is now AI Agent ready ✅ Imagine failing this test in 2026 Two free tools to grade your own site: → IsItAgentReady.com (built by Cloudflare) → AcceptMarkdown.com

🚨 Anthropic's CEO: "software engineering will be fully automated in 12 months." two types of people right now: type 1: uses Claude Code with basic prompts. scared about the future. type 2: knows all 35 commands, tricks, and workflows. ships faster than a team of 5. type 1 gets replaced in 12 months. type 2 does the replacing. I wrote a full guide on how to start. zero experience needed.

@kumarmanish9 @grok please verify it

“Ab Nahi Aaunga, Bata Dena” ? ( I won’t come again, tell them), The pain, anguish & helplessness in his words. This is from Surat Railway station.

@tariqtramboo Where is India standing

Iran has second highest Intelligent people in the world. The international IQ test has placed Iran in second place in its 2025 ranking with an average intelligence quotient (IQ) of 106.63 ~ And that is the main reason for sanctioning Iran 🇮🇷 The only Islamic country that is capable of surpassing the technological capabilities necessary to be a global power.

@EvanLuthra Netflix: ‘Are you still watching?’ Claude: ‘Are you still learning?’ 😄

INSTEAD OF WATCHING NETFLIX TONIGHT, WATCH THIS 1 HOUR FULL CLAUDE COURSE. THANK ME LATER!!!

Steal these 100 Claude Tips from me right now! Comment "Claude" if you want to receive the HD PDF in your inbox

@Mishi_2210 Its 127. 7+7+1 =15 15+15+1=31 31+31+1=63 63+63+1 =127

This Claude Code project setup guide literally makes Claude Code 10x better🤯

@zoo_bear Eid Mubarak bhai

Eid Mubarak everyone. ♥️🫂 #EidUlFitr

Congratulate all Muslims of the world on the occasion of Eid al-Fitr.

Best wishes on Eid-ul-Fitr. May this day further brotherhood and kindness all around. May everyone be happy and healthy. Eid Mubarak!

@denicmarko Perthking.com

@heyblake perthking.com

Drop your project URL Let’s drive some traffic

@UiSavior Option A

A or B?

@IndianGems_ Indian has 🔔

USA has ChatGPT USA has Grok USA has Claude USA has Gemini USA has Llama USA has Copilot China has DeepSeek China has Qwen China has Ernie China has GLM China has Kimi India has?

Safety is designed, not accidental. Read: blog.webgrapple.com/prompt-enginee… #PromptEngineering #SoftwareEngineering #ResponsibleAI #ProductionCode

Most production bugs aren’t careless. They’re written under pressure. Article #3: Prompt Engineering for Writing Safer Code Under Pressure. • Surface assumptions • Enumerate failure modes • Stress-test logic • Keep AI as assistant — not authority