Dmitry Noranovich

Machine Learning Engineer. I learn by teaching. Opinions are my own.

Building

[Andrew Ng @AndrewYNg]

New course on serving LLMs efficiently -- how do you serve models to many concurrent users at low latency and reasonable cost? This short course is built with @RedHat and taught by @cedricclyburn.

Efficient LLM serving requires efficient memory management. A 70B-parameter model takes ~140 GB just to load the weights. On top of that, every active request needs its own chunk of GPU memory, the KV cache, to store the token context it has built up so far. In this course, you'll learn to reduce a model's memory footprint with quantization and serve it using vLLM, which handles many concurrent requests efficiently through smart memory management.

Skills you'll gain:
- Quantize a model and measure the accuracy tradeoff
- Serve a model with vLLM and watch it handle concurrent requests efficiently
- Benchmark your deployment and make informed tradeoffs between speed, cost, and accuracy

Join and learn to serve LLMs efficiently:
https://t.co/x04xMbFlkO

[OpenAI @OpenAI]

We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time.

Today, that work is rolling out as a more capable memory system in ChatGPT. https://t.co/0MyFKCe2Mu

[Sebastian Raschka @rasbt]

And another open-weight release. Nemotron 3 Ultra has an ultra impressive capability:efficiency ratio!

Design-wise, it carries forward the Mamba-2-attention hybrid stack and LatentMoE introduced in the previous Super variant. But everything is a bit bigger.

[Dmitry Noranovich @javaeeeee1]

AppWizzy: Rent a private VM with Codex to build production apps by @flatlogic and @Blarior https://t.co/Jj2hMFbvjY

[Dmitry Noranovich @javaeeeee1]

Google Gemma 4 12B: Run multimodal AI locally with an encoder-free architecture by @joshtwoodward https://t.co/vursWytAwr

[Dmitry Noranovich @javaeeeee1]

Astra Security: AI-Powered Offensive Pentest Platform by @shikhilsharma https://t.co/o17zqzokZy

[Dmitry Noranovich @javaeeeee1]

Carbon Voice: Record your voice. Share a link. Skip the meeting. by @TravisBogard https://t.co/pXhK3F5k3E

[Dmitry Noranovich @javaeeeee1]

Close: This SMB CRM calls your leads for you by @Steli, @NickPersico, and @philfreo https://t.co/m8kGWAhTMb

[Dmitry Noranovich @javaeeeee1]

Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectories

https://t.co/qou8E8vTnb

[Philipp Schmid @_philschmid]

We made a collection @GoogleDeepMind scientific agent skils for research tasks, genomics, structural biology, cheminformatics, literature search, and more.

👉https://t.co/zkPuCtmwEE
https://t.co/zkPuCtmwEE

[DeepLearning.AI @DeepLearningAI]

New short course: Fast & Efficient LLM Inference with vLLM, built in partnership with @RedHat and taught by @cedricclyburn.

Learn to quantize an open-source LLM, serve it with vLLM, and benchmark your deployment across speed, cost, and accuracy.

Free to enroll: https://t.co/czVwJBnLZ6

[Sebastian Raschka @rasbt]

It's been a while! 4 nice additions to the open-weight local-LLM-on-consumer-hardware ecosystem:

[Google Gemma @googlegemma]

Meet Gemma 4 12B!

A unified, encoder-free multimodal model designed to bring high-performance intelligence directly to your laptop, and released under an Apache 2.0 license.

Bridging the gap between edge efficiency and advanced reasoning. Here is what’s new with Gemma 4 12B: 👇

[Dmitry Noranovich @javaeeeee1]

https://t.co/34W7dzliHd

[Dmitry Noranovich @javaeeeee1]

The 2026 AI GPU Buyer's Guide: Benchmarking Datacenter to Budget Hardware

https://t.co/m9dc7McMHv

[Dmitry Noranovich @javaeeeee1]

https://t.co/q2nUoF1dDK

[Dmitry Noranovich @javaeeeee1]

SurrealDB: Storage, context, and memory. One database, one transaction. by @tobiemh, @jaimemh, and @itsezc https://t.co/FGaTYvymNN

[Dmitry Noranovich @javaeeeee1]

Brandfetch: The brand identity platform by @NuriKasimov and @aminkasimov https://t.co/WEOS0DYklU

[Dmitry Noranovich @javaeeeee1]

Carbone: Generate and convert any document efficiently by @carbone_io https://t.co/YRnGfhpHnZ

[Dmitry Noranovich @javaeeeee1]

Handler: Review AI edits like stacked PRs at generation time. by @Vignesh_warar https://t.co/7eF3gaM0hG