Hafeez Al-Haqq

@TheAhmadOsman Alternative? x.com/i/status/20506…

Genuine advice If you need ANY hardware, BUY IT NOW - Phones - Laptops - Computer parts Hardware prices are about to get ridiculous I just bought my wife a new MacBook & iPhone I’m not trying to flex, just getting ahead of the supply shock before the prices get wild

@1weiho Can you add support for dynamic slide (e.g. for data analytics slides) using ReporteRs Package and R Markdown?

Introducing open-slide - The slide framework built for agents. Prompt your agent, get a polished deck. $ npx @⁠open-slide/cli init 👇

30 agents every AI Engineer must build. This is the most comprehensive and practical book on AI Engineering that I've ever seen. I can't think of a single use case that they didn't cover here: 1. The autonomous decision-making agent 2. The planning agent 3. The memory-augmented agent 4. The knowledge retrieval agent 5. The document intelligence agent 6. The scientific research agent 7. The tool-using agent 8. The agentic workflow system 9. The data analysis agent 10. The verification and validation agent 11. The general problem solver agent 12. The code generation agent 13. The security-hardened agent 14. The self-improving agent 15. The conversational agent 16. The content creation agent 17. The recommendation agent 18. The vision language agent 19. The audio processing agent 20. The physical world sensing agent 21. The ethical reasoning agent 22. The explainable agent 23. The healthcare intelligence agent 24. The scientific discovery agent 25. The financial advisory agent 26. The legal intelligence agent 27. The education intelligence agent 28. The collective intelligence agent 29. The embodied intelligence agent 30. The domain-transforming integration agent I also read 50 Algorithms Every Programmer Should Know by Imran. Same vibe. Here is the Amazon link: amzn.to/4t5ystE

Lego Movie: Mayday Distress

Iran defeating 50 years of U.S. propaganda with Lego videos is the one bright spot of 2026.

🇮🇷 The leader of the Jewish community in Iran "pledged allegiance" to the third Leader of the Iranian Revolution and the Republic of Iran, Sayyed Mojtaba Khamenei, praying, "He succeeds in carrying out this great responsibility with honor and distinction."

This is actually insane. 🤯 Claude Opus 4.6 has around 200B active parameters and registers 75% on SWE-bench verified. Meanwhile, Qwen 3.6 35B A3B has 3B active parameters and scores 73.4% on SWE-bench verified. Same benchmark, but Qwen has 60x fewer active parameters.

@jahirsheikh8 Do you have any tips or practical resources on how to enable cache context to save tokens on OpenClaw? Something like using WebSocket?

You’re not an AI Engineer until you understand these terms: • 🧠 Embeddings → Numerical meaning of text/data • 🔍 Vector DB → Similarity search storage • 📚 RAG → Retrieval-Augmented Generation • 🎯 Fine-Tuning → Task-specific model training • 🪶 LoRA → Lightweight fine-tuning method • ⚡ Quantization → Smaller/faster models • 🧵 Context Window → Model memory limit • 🔄 Function Calling → Structured tool usage • 🛡 Guardrails → Output constraints/safety • 📏 Eval Frameworks → Measure model quality • 🧮 Tokenization → How text becomes tokens • 🚀 KV Cache → Faster inference reuse • 🔥 Hallucination → Confident wrong output • 🪝 Prompt Chaining → Multi-step workflows Building demos is easy. Production AI is not.

@bindureddy Have you coupled it with @CommandCodeAI better result + more savings.

Finally figured out how to run DeepSeek V4 Pro correctly The evals are mind blowing... Yes, it is even better than Opus 4.7 Medium!! More coming...

@simplifyinAI Cc: @akmalalif0v0

Someone just open-sourced shadcn-style map components for React. It's called mapcn. Theme-aware, fully composable, and built on MapLibre GL so you get full mapping power without the Mapbox bill. 100% Open Source.

Excited to announce our first video course Introduction to Command Code - 9 lessons. From scratch to shipping real apps. - Build software that matches your coding taste. - Live on our website and YouTube. Start for free ↳commandcode.ai/courses

Extremist is the death and destruction that Benjamin Netanyahu government brings on a daily basis. Extremist is them introducing the death penalty for Palestinians. Extremist is this apartheid, genocidal Israeli Government that our Government shamefully supports.

@drhossamsamy65 @ArnoudvDoorn H💚🍉, 🇵🇸🤍 E

🚨Spain🇪🇸, Ireland🇮🇪, Slovenia🇸🇮, Belgium🇧🇪, Finland🇫🇮, France🇫🇷, and Sweden🇸🇪 will tomorrow submit a formal request to the European Union to terminate the partnership agreement with #Israel.

@amasad @Replit Peak dogfooding in the history of mankind. Kudos @Replit

Shout out to @replit engineers and support team keeping everything together as users run armies of agents building everything they ever dreamed of 😅

@DataChaz I've heard about active context vs cache context, but still unclear how to practically implement those. Any tip?

🚨 Let's look at the API math behind AI agents. Someone just ran 831,962,136 tokens through DeepSeek for only $10.57. When building tools like OpenCode, output tokens dictate your burn rate because the agent must constantly re-read your codebase. If we migrate those 831.9M tokens over to Anthropic’s Opus 4.7 ($5/M input, $25/M output), here is the damage based on your I/O split: > 100/0 Split (Pure context, unrealistic): $4,159.81 > 90/10 Split (Standard coding agent): $3,743.83 (in) + $2,079.91 (out) = $5,823.74 > 80/20 Split (Higher output tasks): $3,327.85 (in) + $4,159.81 (out) = $7,487.66 > 50/50 Split (Maximum output cost): $2,079.90 (in) + $10,399.53 (out) = $12,479.43 You'd be dropping $5k to $8k on Claude for a workload that costs less than lunch on DeepSeek. A 500-700x price drop for similar-ish performance is a wake-up call for the entire industry.

Swedish activist Greta Thunberg says @gbsumudflotilla members are being tortured by the israelis: we must demand their immediate release

@keennay Good stuff, but why is it still hard to find peeps that experimenting with AMD RDNA V3.5-4? I'm curious about ROCm 7 vs CUDA.

New experiment: - Qwen3.5-122B-A10B-FP8 (127GB) - Inference Engine: KTransformers + SGLang - Hybrid GPU + CPU memory offload across 1x RTX Pro 6000 (96GB GDDR7) & 121GB DDR5 RAM - 176 out of 256 routed experts per MoE layer are on the GPU

@MrAhmadAwais @jbhuang0604 @CommandCodeAI nxcode.io/resources/news…

@MrAhmadAwais @jbhuang0604 @CommandCodeAI So vanilla deepseek v4 is 50x cheaper than opus, so adding @CommandCodeAI spikes another 60x savings on top 🤯

Keep getting rate-limited by Claude, so I tried out DeepSeek V4 for the first time. After 10M+ tokens, holy crap the cost is ... 🤯