Live feed

Pandaily · 2026-11-14 · Ranked: community signal · fresh 0.98 · score 2.87 · Context

DeepSeek Opens Its App's Search, File, OCR and Voice Tools as APIs So Developers Can Rebuild the App

kingstonyip.com · 2026-11-14 · Ranked: agent match · community signal · fresh 0.97 · score 2.78 · Context

Show HN: Portal2Bench – Can AI agents solve Portal 2?

blog.cloudflare.com · 2026-11-13 · Ranked: community signal · fresh 0.94 · score 2.66 · Context

Adapting to AI penetration testing orchestrators: The Modern Security Model

infoq.com · 2026-11-12 · Ranked: agent match · practitioner analysis · fresh 0.69 · score 2.30

Presentation: Opening the Black Box: Critical Explainability for AI Agent Tool Selection with Kiji Inspector

Hannes Hapke shares how Dataiku’s open-source Kiji Inspector applies Sparse Autoencoders (SAEs) directly to decision tokens to expose why agents pick specific tools. Moving past superficial chain-of-thought narratives... Context & related coverage →

vllm.ai · 2026-11-12 · Ranked: agent match · practitioner analysis · fresh 0.64 · score 2.26

DeepSeek-V4.1-Flash on Vera Rubin NVL72: 58x Throughput vs. GB300 NVL72 at 387 tokens/s

On one NVIDIA Vera Rubin NVL72 rack, vLLM serves DeepSeek-V4.1-Flash on SemiAnalysis AgentX at 1.7M tokens/s per GPU at a P90 of 387 tokens/s per user, 58x GB300 NVL72 at the same interactivity. Context & related coverage →

arxiv.org · 2026-11-12 · Ranked: research watch · fresh 0.81 · score 2.26

Shortcuts or Memorisation? Revisiting Data Corruption in Natural Language Understanding with Large Language Models

Pre-trained encoders fine-tuned on natural language inference (NLI) and other language understanding tasks keep much of their accuracy after the input is perturbed, e.g. by removing all nouns or shuffling the words. D... Context & related coverage →

arxiv.org · 2026-11-12 · Ranked: evaluation match · research watch · fresh 0.77 · score 2.26

SAGE-QA: Semantic-Aware, Grounded Evaluation for Closed-Context Question Answering on SQuAD 2.0

Closed-context question answering (QA) requires a system to generate an answer strictly supported by a supplied passage, or to abstain when the passage lacks adequate evidence. SQuAD 2.0 is a widely used benchmark for... Context & related coverage →

blog.cloudflare.com · 2026-11-13 · Ranked: agent match · cloud platform update · fresh 0.91 · score 2.21

Kitesurf is now open source

Kitesurf, our experimental browser engine for AI agents, is now open source. Inspect the code, run it locally, or deploy your own instance on Cloudflare Workers. Context & related coverage →

deepmind.google · 2026-11-12 · Ranked: frontier engineering · fresh 0.82 · score 1.95 · Context

Introducing Gemini 4 Hydrogen

vllm.ai · 2026-11-12 · Ranked: practitioner analysis · fresh 0.64 · score 1.70

Inkling 2 Small on vLLM: Day-0 Support

vLLM supports Thinking Machines' Inkling 2 Small from day 0: attention residuals, a new decode kernel for relative-bias attention, all-reduce fused with the ops around it, and DSpark speculative decoding. Context & related coverage →

openai.com · 2026-11-13 · Ranked: agent match · frontier engineering · fresh 0.89 · score 1.66

Shield helps technicians hand off legacy portal work with OpenAI

Shield uses OpenAI to help IT technicians hand off work in legacy portals. Its Forge agents navigate systems so technicians can focus on issues needing their judgment. Context & related coverage →

openai.com · 2026-11-12 · Ranked: agent match · frontier engineering · fresh 0.83 · score 1.55

How Amplitude ships product improvements in 1–2 days with OpenAI

Built primarily on OpenAI models, Amplitude’s proactive product agent, Wave, identifies what to build, helps teams ship changes with coding agents, and measures the impact. Context & related coverage →