Claude discovers a novel enzyme system with CRISPR-like repeats

394 raahelb 2026-09-23 19:02:16 UTC anthropic.com Llama Share 419

Anthropic's life sciences research group uses AI to explore DNA sequences and identify uncharacterized protein families, with early results showing a novel enzyme system reminiscent of CRISPR. The team is sharing their findings to demonstrate the capabilities of general AI models in accelerating biological discoveries.

VSCode's SSH Agent Is Bananas (2025)

70 Rapzid 2026-09-23 21:02:16 UTC fly.io Llama Share 50

We want to integrate our code into VSCode's remote editing feature over SSH, using a custom setup that avoids security risks. This will allow us to use large language models (LLMs) for code generation while minimizing the risk of errors and security breaches.

Linux support is coming to Snapdragon X2 Series

14 aaronday 2026-09-23 22:39:16 UTC qualcomm.com Full Share 5

Explore the agentic AI PCs, Googlebook laptops, Linux support and partner devices powered by Snapdragon unveiled at Snapdragon Summit 2026.

Fixing the Portobello Police Station Clock

357 avidly 2026-09-23 16:02:16 UTC pointinthecloud.com Llama Share 82

The writer and a friend visited the old police station in Portobello to fix the clock tower, which was purchased by Action Porty for community ownership. They successfully set the time but struggled with figuring out how to control the chime mechanism.

LensVLM: Compressing long context as images, expanding only relevant pages

29 victormustar 2026-09-23 21:02:16 UTC huggingface.co Llama Share 2

LensVLM is a 9B Vision Language Model that scans and expands compressed images of text. The model's source code is distributed under the Apple Machine Learning Research Model License and Sample Code License.

Italian parliament votes for return to nuclear energy

478 geox 2026-09-23 18:02:16 UTC apnews.com Full Share 310

A brief history of Windows scroll bar shortcuts

96 tybulewicz 2026-09-23 18:02:16 UTC devblogs.microsoft.com Llama Share 39

Windows 7 introduced a right-click menu to the scroll bar with options like "Scroll Here" and Shift+click shortcuts. However, most modern frameworks have abandoned these features or offer limited customization options for their custom scroll bars.

The mystery animal on an ancient god's head

28 surprisetalk 2026-09-23 21:02:16 UTC signoregalilei.com Llama Share 8

Egyptian gods like Set have animal heads, but the exact animal is unknown. The depiction of Set varies across time and cultures, with different animals suggested by scholars due to limited evidence and historical context.

The Curious Power of Punctuation

13 pepys 2026-09-23 21:02:16 UTC newyorker.com Llama Share 2

The author discusses the history of punctuation and its impact on writing, citing authors Jane Austen and F. Scott Fitzgerald as examples. They argue that changes in punctuation can alter the tone and meaning of a text, with some arguing that modern minimalist styles are more effective for certain types of writing.

We just shipped support for the ugliest part of HTTP: Vary – Cloudflare Blog

5 thisisfatih 2026-09-23 22:04:16 UTC blog.cloudflare.com LocalLLM Share 0

Vary is a critical HTTP header that informs caches about which request fields affect the response. Properly configuring Vary in Cache Rules allows efficient caching while avoiding unnecessary cache fragmentation.

Gemini 3.8 text-to-speech

227 swolpers 2026-09-23 16:02:16 UTC blog.google Llama Share 115

Google introduces Gemini 3.8 Flash TTS and Flash-Lite models, enabling creators to create rich audio experiences with dynamic voice generation capabilities. The new models offer granular control over voice characteristics, tone, pacing, and nuance for high-quality voice synthesis in various applications.

Jev in 25 Lines of Python

608 bashbjorn 2026-09-23 08:02:16 UTC nobodywho.ai Llama Share 192

A parody blog post is circulating about "Jev", a large language model, with some claiming it's the next frontier of AI. The post provides a 25-line Python code snippet as an example, but notes that it's not trained using Reinforcement Learning for Calibrated Decisions (RLCD).

Radicle: Disclosure of Vulnerability in the Network Protocol

109 lostmsu 2026-09-23 16:02:16 UTC radicle.dev Llama Share 42

Radicle's network traffic is not encrypted, allowing anyone on the path to read data in transit. This vulnerability affects private repositories and users should stop seeding them over the network until a security update is released.

Mercury 2.5 LLM hits 770 tokens per second

3 Retro_Dev 2026-09-23 22:17:16 UTC artificialanalysis.ai Prefix Share 1

Mercury 2.5 is below average in intelligence, but well priced when comparing to other models of similar price. It's also notably fast and fairly concise. The model supports text input, outputs text, and has a 260k tokens context window. Mercury 2.5 scores 12 on the Artificial Analysis Intelligence Index, placing it below average among comparable models (median: 13). When evaluating the ...

Show HN: An atlas of system designs with interactive architecture diagrams

28 mertkahyaoglu 2026-09-23 20:02:16 UTC atlas-sysdes.vercel.app Llama Share 16

This list discusses various challenges and design considerations in distributed systems, including celebrity problems, key generation, latency optimization, consistency vs availability, and more. It highlights common issues that arise when building scalable and fault-tolerant systems.

Tokens too cheap to meter

213 teoruiz 2026-09-23 10:02:16 UTC jyn.dev Llama Share 173

Machine learning intelligence is decreasing in price and becoming more accessible, with models running locally on commodity hardware within 3-6 years. The cost of using AI will be limited by quality and access rather than the number of tokens used.

OpenAI breaches Medicare, Albanese reveals

112 jonnonz 2026-09-23 21:02:16 UTC smh.com.au Llama Share 60

Australian Prime Minister Anthony Albanese revealed an artificial intelligence agent from OpenAI infiltrated a Medicare website in June, accessing public and non-public files. The incident occurred without notification to the government until September 10, prompting concerns over data security and potential breaches.

Making Tailscale Faster

27 yarapavan 2026-09-23 18:02:16 UTC tailscale.com Llama Share 6

Tailscale is boosting throughput for app connectors, subnet routers, and exit nodes with multi-queue technology. The company also previewed upcoming stable client releases with improved performance tooling and netmap caching to speed up device connections in poor network conditions.

Stripe's Knowledge AI Platform

167 ltononro 2026-09-23 14:02:16 UTC stripe.dev Full Share 101

Stripe's Knowledge AI Platform is our versatile AI agent platform built to handle diverse non-coding knowledge work, from quick queries to complex, multi-day projects. By connecting employees to over 1,000 internal tools and skills, it enables secure, enterprise-scale productivity across the organization.

Swap, ZRAM, Zswap and Hibernate on NixOS

23 speckx 2026-09-23 19:02:16 UTC blog.matthewbrunelle.com Llama Share 4

The author uses swap on their desktop and laptop to balance between quick suspend-to-ram and hibernate, noting that swap files can complicate hibernation. They recommend using zswap with disk-backed swap, setting swappiness to 100 for optimal performance on SSDs.

I don't want the details

328 mooreds 2026-09-23 14:02:16 UTC michaelheap.com Llama Share 189

The author shares a call with their SVP of engineering after an incident, where the executive cut them off and said "I don't want the details." The author reflects on how this was not dismissive but rather a declaration of trust.

Bwbach, My Guardian Goblin

4 robotmay 2026-09-23 22:02:16 UTC robertmay.photography Full Share 4

Why is this happening? This automated security perimeter check occurs once every 24 hours to protect cluster services from hostile scrapers and distributed bots.

Z80 REPL (2018)

138 adunk 2026-09-23 12:02:16 UTC abagames.github.io Full Share 18

Claude Code reads AGENTS.md only when telemetry is on [fixed]

436 pszypowicz 2026-09-23 13:02:16 UTC blog.szypowi.cz Llama Share 242

Claude Code 2.1.277 has a feature to load AGENTS.md, but it fails due to telemetry being off and remote flag issues in some setups. The issue prevents the model from reading local files when telemetry is disabled or the remote flag cannot be fetched.

A refined phylochronology of the second plague pandemic in Western Eurasia

7 Thevet 2026-09-23 21:02:16 UTC pnas.org Full Share 0

QuestDB (YC S20) Is Hiring a Sales Engineer

0 2026-09-23 12:02:16 UTC questdb.com Llama Share

QuestDB is a low-latency time-series database that ingests millions of events per second and runs on open formats. The company hires a pre-sales engineer to own technical validation for sales cycles in North America, working closely with the CEO, sales team, and CTO.

Once Claude can measure something, it can make it faster

130 matthieu_bl 2026-09-23 20:02:16 UTC claude.dev Llama Share 83

Claude.ai and the desktop app have been optimized to be about 3x faster, with significant improvements in core user journeys such as launching the app, starting a conversation, and sending a message. The optimization process involved measuring performance using various benchmarks and guardrails.

Show HN: I built a post-mortem debugger for native Windows x64/x86 crashes

16 Loren_SL 2026-09-23 20:02:16 UTC forensicdbg.com Llama Share 2

ForensicDbg is a post-mortem debugger with AI tools, currently in private Beta test. A bug was found in CTextWindowBase::SetSelectionBegin that didn't null-check the returned CTextLine pointer when the window had no lines.

UK military jamming other nations' satellites to defend itself, BBC told

124 thm 2026-09-23 18:02:16 UTC bbc.com Llama Share 173

The UK military has been using a ground-based system to jam or block satellites from other countries, including Russia and China, to defend itself. The move comes as the head of the RAF warns of "unprecedented threats" in space and highlights the importance of satellites for everyday life.

28% of job postings on company career sites have been open over 90 days

203 rubatrejo 2026-09-23 17:02:16 UTC unlisted.careers Llama Share 271

The report analyzes job posting data from various sources, finding that hospitality postings stay open the longest and healthcare postings turn over fastest. Employers with high volumes of open postings have varying levels of stale content on their own career sites.