Microsoft Corp. today released software updates to plug at least 570 security holes in its Windows operating systems and other software, almost triple the number of vulnerabilities the software giant ...

The relational database, redesigned. With one cohesive toolchain. From schema to application.

Dependabot now waits until a new release has been available on its registry for at least three days before opening a version update pull request. This cooldown is now the…

Catches silent CPU fallback and mislabeled tok/s in local LLMs. llama.cpp and ollama, one file, no deps. - logxio/picchio

This paper reveals and exploits a critical security vulnerability: the electromagnetic (EM) side channel of capacitive touchscreens leaks sufficient information to recover fine-grained, continuous han...

We ran autoresearch on autoresearch: an outer loop rewriting its own research agent for 100 unattended steps. In eight days it discovered agents that beat our two years of hand-tuning on held-out benc...

Packing ternary numbers with efficient SIMD-friendly unpacking on binary computers

For decades, one of the most notoriously-challenging problems on the web has been sticking one element to another element, for things like tooltips and nested menus. The CSSWG has decided to provide a...

Pebble Mega Update - July 2026

We tracked ChatGPT and Claude connectors for 6 weeks and found that a connector changes every nine minutes: gaining write access, widening its permissions, modifying its toolset, and more.

Open models have reached impressive levels of capability, but the surrounding platform often falls short. This blog explores the hidden infrastructure gaps developers face and why bridging them is ess...

A unified, unmanaged multimodal runtime. Zero-allocation native AI inference for Java 22+. - ProjectArgus-cc/libargus.cc

Find out about the new features in .NET 11 Preview 6 across runtime, SDK, libraries, ASP.NET Core, .NET MAUI, C#, Entity Framework Core, F#, and container images.

An open-source, authenticated pull-through cache for Hugging Face. - pulsys-io/pulsys

Very low latency speech to text, intent recognition, and text to speech, for building voice agents and interfaces - moonshine-ai/moonshine

A cross-platform desktop IRC client for the modern era complete with scripting and plugins - matt0x6F/irc-client

Advanced glycation end products (AGEs) in proteins, a hallmark of aging, are considered irreversible. Here, authors report the development of CMLase - an enzyme that specifically oxidizes Nε-carboxyme...

Next-gen container runtime with zero-copy DAG storage and P2P image sync. Run OCI images as Linux containers, Firecracker microVMs, or Apple Silicon VMs. - pullrun/pullrun

The Free and Open Source 3D Creation Software — blender.org

Self host your own server for Atuin AI - the agent that lives in your terminal, and knows how you work

Self host your own server for Atuin AI - the agent that lives in your terminal, and knows how you work

The Free and Open Source 3D Creation Software — blender.org

Reducing reserved memory on NVIDIA GPUs. Contribute to lmganon16/nvidia-vram-research development by creating an account on GitHub.

The vulnerability nobody seems interested in fixing

The concept of recursive self-improvement (RSI) dates back to I. J. Good (1965), where he defined an “ultraintelligent machine” as a system that can surpass humans in all intellectual activities and d...

We added copy-on-write image layering inside Firecracker's block device, and the disk got faster, not slower. Snapshots pause the VM for tens of milliseconds and cost only what changed, so a running P...

A deep dive into all-gather, reduce-scatter, all-reduce, and all-to-all communication on TPUs and GPUs.

S&P Global has downgraded Oracle's rating due to massive AI investments. The company is now only one notch away from the speculative territory.

Capacitive touchscreens in modern smartphones introduce severe side-channel vulnerabilities. However, existing attacks often require restrictive conditions or invasive measurements. This paper present...

Vibecoding and the possible collapse of a shared language.

I built a device to measure end-to-end input latency, then used it to find out what actually moves the needle.

We introduce Gemma 4, a new generation of open-weight, natively multimodal language models in the Gemma model family. Designed to advance compute efficiency and reasoning, the Gemma 4 model suite feat...

Leaderboard scores don’t hold up when we look more closely at benchmarks that are considered standards. The models dramatically improved on the exact tasks we trained them on, yet those gains often di...

Observability for AI-Native teams. Logs, metrics, traces, APM and LLMOps — drop-in replacement for Datadog, Langfuse, and Grafana at 1/5th the cost.

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Reflections on autonomy and the value of thinking for ourselves

Twenty-six former employees of Meta Platforms have filed a ​lawsuit against the tech company, accusing it ‌of using AI-powered software that disproportionately targeted people with disabilities or who...

Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.

ArcadeDB 26.7.2 ships full Bolt driver certification: 39 conformance scenarios run nightly against the Java, JavaScript, Python, .NET, and Go Neo4j drivers across 14 pinned versions, published as a li...

The first open-source modern Apple Silicon emulator.

New York became the first U.S. state on Tuesday to halt construction of large new data centers, imposing a ‌one-year moratorium as concerns grow that the facilities driving the artificial-intelligence...

OpenAI mandates hardware-backed passkeys for Trusted Access Cyber (TAC) members to log into ChatGPT accounts

The Juggler Code Agent. Contribute to juggler-ai/juggler development by creating an account on GitHub.

Build real-time interactive AI avatars with Anam's API. Low-latency conversational video AI for customer service, training, and sales.

How should we work with LLMs when growing abstractions?

AI applications are driving explosive data growth and collapsing the line between transactional and analytical workloads, making the best-of-breed combination of Postgres and ClickHouse — increasingly...