Md Rahmat Ullah

I architect next‑gen AI pipelines that reduce inference costs by a good margin while scaling generative AI and large language model applications.

[Md Rahmat Ullah @md80182]

GraphBit: Developer-first, enterprise-grade LLM framework. https://t.co/1RGmcSUj3F by @md80182

[Md Rahmat Ullah @md80182]

GraphBit: Developer-first, enterprise-grade LLM framework. https://t.co/1RGmcSUj3F by @md80182

[Md Rahmat Ullah @md80182]

My thoughts on GraphBit on @ProductHunt cc @GraphBit https://t.co/bukOwwABMV

[Md Rahmat Ullah @md80182]

GraphBit: Developer-first, enterprise-grade LLM framework. https://t.co/1RGmcSUj3F by @md80182

[Md Rahmat Ullah @md80182]

GraphBit is hosting Launching soon: The Ultra-Efficient Enterprise Grade Agentic AI Framework. Make sure to attend it today. https://t.co/y0yShNOlk0

[Md Rahmat Ullah @md80182]

Most agent frameworks look great in demos… then fall apart at step 3 of 10. We built GraphBit to change that—a declarative framework for reliable, observable, multi-step AI workflows that actually ship.

[Md Rahmat Ullah @md80182]

ANN isn’t a library choice — it’s an operating point. If you’re shipping vector search, pick your method to match metric, latency/recall, and memory, not the leaderboard.

https://t.co/VB0pqsuOA6

[Md Rahmat Ullah @md80182]

GraphBit: Developer-first, enterprise-grade LLM framework. https://t.co/1RGmcSUj3F by @md80182

[Md Rahmat Ullah @md80182]

Grok 2.5 model, which was xai's best model last year, is now open source.

Grok 3 will be made open source in about 6 months.

[Elon Musk @elonmusk]

The @xAI Grok 2.5 model, which was our best model last year, is now open source.

Grok 3 will be made open source in about 6 months.

https://t.co/TXM0wyJKOh

[Elon Musk @elonmusk]

The @xAI Grok 2.5 model, which was our best model last year, is now open source.

Grok 3 will be made open source in about 6 months.

https://t.co/TXM0wyJKOh

[Md Rahmat Ullah @md80182]

“Don’t even bother with a law or medical degree?”
A founder of Google’s first Generative AI team argues that by the time you graduate, AI will have reshaped those careers beyond recognition, and advises aiming for niche intersections like AI × biology instead.

[Md Rahmat Ullah @md80182]

AI at a Crossroads: For People, Not as People

In August 2025, Mustafa Suleyman (Head of Microsoft AI) raised an urgent reminder:

“We must build AI for people, not to be a digital person.”

#AI #Ethics #ResponsibleAI #AgenticAI #FutureOfAI

[Md Rahmat Ullah @md80182]

🔥 @OpenAI just dropped GPT-OSS: 120B & 20B open-weight LLMs under Apache-2.0! 128k ctx, full chain-of-thought, MoE. 120B serves on a single H100; 20B runs on a laptop (16 GB). Near-parity with o4-mini. Privacy-friendly, hackable, free. 👇
🔗 https://t.co/rV4PNRN9WL
#AI #LLM

[Anthropic @AnthropicAI]

Today we're releasing Claude Opus 4.1, an upgrade to Claude Opus 4 on agentic tasks, real-world coding, and reasoning.

[OpenAI @OpenAI]

Our open models are here.
Both of them.

https://t.co/9tFxefOXcg

[Md Rahmat Ullah @md80182]

@huggingface @OpenAI Finally, we are getting something to test.

[Md Rahmat Ullah @md80182]

Tweet:
@AnthropicAI Claude Opus 4.1 is live! 🚀
• 74.5 % on SWE-bench Verified, new SOTA coding 🏆
• Sharper agentic search & deeper reasoning
• Precise multi-file refactors devs love

Now on Claude Code, API, AWS Bedrock & GCP Vertex, same price. Try it!
#AI #LLM #DevTools

[Md Rahmat Ullah @md80182]

🚀 5 new open-source #LLM in July 25 (Chatbot Arena Elo — 1 Aug):

1️⃣ Qwen3-235B A22B Inst 1479
2️⃣ Kimi K2-0711 1459
3️⃣ Qwen3-235B A22B Think 1446
4️⃣ Qwen3-Coder-480B 1432
5️⃣ GLM-4.5 ~1330*
Long-context & trillion-param MoE models are here. #OpenSource #AI

[Md Rahmat Ullah @md80182]

I just published Probabilistic Forecasting Techniques https://t.co/EyQ7A4D1ti

[Md Rahmat Ullah @md80182]

I just published Why I Moved My Llama-3 Workloads to vLLM and Why You’ll Probably End Up Doing the Same https://t.co/WDpWFe0Bpu