When you build AI agents, don't treat prompts like config strings. Treat them like executable business logic. Because that's what they really are. @arshdilbagi's blog and this Stanford CS 224G lecture lay out one of the clearest mental models I have seen for LLM evaluation. Stop treating evals like unit tests. That works for deterministic software. For LLM products, it creates false confidence because real-world usage changes over time. Example: an insurance prompt passed 20 eval cases. The team shipped. In production, a new class of requests showed up and failed quietly. No crash, no alert, just wrong answers at scale. The fix is not "write more eval cases," which is what many teams do. It is building evals as a living feedback loop. Start with a small set, ship, watch what breaks in production, add those failures back, and re-run on every prompt or model change. What eval failure caught your team off guard? Blog: https://t.co/HCVhcow5rA Stanford CS 224G lecture: https://t.co/q667gGwckt

When you build AI agents, don't treat prompts like config strings. Treat them like executable business logic. Because that's what they really are. @arshdilbagi's blog and this Stanford CS 224G lecture lay out one of the clearest mental models I have seen for LLM evaluation. Stop treating evals like unit tests. That works for deterministic software. For LLM products, it creates false confidence because real-world usage changes over time. Example: an insurance prompt passed 20 eval cases. The team shipped. In production, a new class of requests showed up and failed quietly. No crash, no alert, just wrong answers at scale. The fix is not "write more eval cases," which is what many teams do. It is building evals as a living feedback loop. Start with a small set, ship, watch what breaks in production, add those failures back, and re-run on every prompt or model change. What eval failure caught your team off guard? Blog: https://t.co/HCVhcow5rA Stanford CS 224G lecture: https://t.co/q667gGwckt

Google Workspace CLI: https://t.co/Rg229zYsoA

Google Workspace CLI: https://t.co/Rg229zYsoA

Read on for more: https://t.co/bLxqT3yUTl

Read on for more: https://t.co/bLxqT3yUTl

thanks codex team for supporting the radare2 project with some Pro subs!🙏❤️ https://t.co/0fKEt16AeI

thanks codex team for supporting the radare2 project with some Pro subs!🙏❤️ https://t.co/0fKEt16AeI

that was the old me, the six days ago me https://t.co/btAJYAwAjL

amazing work team https://t.co/9d1cEx7vy2

Couldn't help it! Had to give GPT 5.4 (High) + /fast mode a try. → Added height terrains to the level → Animation tweens for the jumps Used xHigh to solve a gnarly bug with the controls successfully 💪 This Final Fantasy Tactics-inspired game was completely vibe coded! https://t.co/q2K7PovU62

We’re announcing Kos-1 Lite, a medical model that achieves SOTA on HealthBench Hard at 46.6%. As a medium sized language model (~100B), it achieves these results at a fraction of the serving cost of frontier trillion-parameter models. https://t.co/27sxAHPgZM

We’re announcing Kos-1 Lite, a medical model that achieves SOTA on HealthBench Hard at 46.6%. As a medium sized language model (~100B), it achieves these results at a fraction of the serving cost of frontier trillion-parameter models. https://t.co/27sxAHPgZM

SWE-rebench V2 A language-agnostic pipeline that automatically harvests 32,000+ executable real-world software engineering tasks across 20 programming languages. Built for large-scale RL training of code agents with reproducible Docker environments. https://t.co/JJ0vLH5N7B

SWE-rebench V2 A language-agnostic pipeline that automatically harvests 32,000+ executable real-world software engineering tasks across 20 programming languages. Built for large-scale RL training of code agents with reproducible Docker environments. https://t.co/JJ0vLH5N7B

Image Generation with a Sphere Encoder https://t.co/6I2FbpogaC

Utonia Toward One Encoder for All Point Clouds paper: https://t.co/AJFPivgBm9 https://t.co/Xbux4iY1QV

BeyondSWE Can Current Code Agent Survive Beyond Single-Repo Bug Fixing? paper: https://t.co/IrLgJJomQU

Beyond Language Modeling An Exploration of Multimodal Pretraining paper: https://t.co/GmtPAQDo8T

Beyond Length Scaling Synergizing Breadth and Depth for Generative Reward Models https://t.co/25QhR93OKK

Kiwi-Edit Versatile Video Editing via Instruction and Reference Guidance https://t.co/s9xlDgXhfc

BBQ-to-Image Numeric Bounding Box and Qolor Control in Large-Scale Text-to-Image Models paper: https://t.co/54U6zmx2ZA https://t.co/fW8zbIrE19

Video world models today have a very limited context length. Mode Seeking meets Mean Seeking (MMM) unlocks long-context, persistent video world models through a unified representation. 1/8 🧵 https://t.co/XXMic82qoc

The Faster-Qwen3-TTS demo just passed the official Qwen3-TTS demo in Hugging Face trending (last 7 days). Now the #5 most trending Space 🚀 https://t.co/8Ar622nKU9

Helios Real Real-Time Long Video Generation Model paper: https://t.co/ae0ZH4zPzn https://t.co/kCnNfF3ImI

Heterogeneous Agent Collaborative Reinforcement Learning https://t.co/ASb1VwtCeK

Proact-VL A Proactive VideoLLM for Real-Time AI Companions https://t.co/GkHdSKxSvi

CubeComposer Spatio-Temporal Autoregressive 4K 360° Video Generation from Perspective Video paper: https://t.co/mnDM1VrYn7 https://t.co/iHtlZJCo1w

LTX-2.3 is out on Hugging Face model: https://t.co/te5nwPL1LE https://t.co/biO7szxFGz

Tencent released HY-WU on Hugging Face An Extensible Functional Neural Memory Framework and An Instantiation in Text-Guided Image Editing model: https://t.co/jAnic8Z9i1 https://t.co/LsLpyjMVQT

New model updates from iquestlab. If you're trying to find an inference model that you can run offline, this is probably the one you're looking for. - 7B and 14B coding models - Optimized for tool use, CLI agents and HTML generation - 128k context length - Explicit and detailed prompting works best - MiT license with requirement of display logo - available on @huggingface

With the help of @huggingface we (/w @RisingSayak) are building a ML Club India 🇮🇳 What we want to do: 1. Online talks 2. IST compatible timing 2. Open to all More to come in this week! Watch this space. 🤗 Special thanks to @LysandreJik who motivated me to keep working on this. 🔥