Tech News

Hacker News readers describe varied AI agent stacks, mixing paid plans like Claude Ultra, Google Ultra, and Z.ai with free OpenRouter models, DeepSeek credits, and local open-weight models. Many build custom orchestrators and subagent frameworks, often wired into Slack or CLI tools, while others rely on GitHub Copilot CLI and Docker/microVM sandboxing for safety. Cost control, token efficiency, transparency, and isolation emerge as key concerns.

OpenAI has built an autonomous pipeline that discovered, validated, and generated patches for 948 local privilege-escalation vulnerabilities in the Linux kernel. The system automates key vulnerability-handling steps from detection through patch creation.

Images 2.6 introduces faster image generation, GIF creation, and more precise editing. The update also adds new API controls and is available across ChatGPT, Codex, and the API.

SpaceX’s Starfall initiative aims to make microgravity access affordable and routine, leveraging the company’s advanced rockets and spacecraft. Founded in 2002, SpaceX seeks to revolutionize space technology with the long-term goal of enabling people to live on other planets.

A new report finds that virtually no DoorDash customers are using AI agents to place their orders, despite growing hype around agentic commerce. The low adoption suggests the e-commerce use case is still in its early days and could evolve as the technology matures.

A Rotherham couple say they have not bought anything in their town centre for 20 years, reflecting how many locals have abandoned UK high streets for retail parks. The piece examines why high streets have declined—down to a mix of vape shops, nail bars and charity shops—and what it would take to lure shoppers back.

XSLT, the longstanding XML transformation standard, is facing retirement as browser vendors move to drop support for it. The change signals the end of a once-common way to transform XML into HTML on the web.

Growing concern over microplastic ingestion is driving a new wave of plastic-free coffee pods and brewing machines. The trend reflects rising consumer demand for safer, more sustainable coffee options.

The article examines the frustrations and reliability problems developers face when working with AI coding agents, focusing mainly on Cursor’s Composer 2.5. It notes that some issues may be model-specific but have also appeared in other models such as GPT-5.5 and Opus 4.6.

The article asks whether curiosity is good for us, focusing on the superego and conscience as the part of us that shows no curiosity about ourselves. It examines how this inner moral voice may limit self-inquiry.

Gauge is a new open-source CLI that recursively improves documentation and developer resources. It can be applied to skills, CLIs, SDKs, MCP, APIs, or docs to iteratively refine them for AI agents.

PageScribe introduces two open-source KOReader plugins: one adds handwriting input for notes and text, while the other provides optional offline dictionary audio pronunciation. Both are designed around everyday reading workflows.

A developer has recreated Apple Vision Pro in Unity, and it runs on both iPhone and Android phones. He also released printable 3D models of a headset that holds your phone, so you can use it much like a Vision Pro. Really an amazing project! 🫶🙌

Indie maker Jack Friks has collected 60+ motivational memes, pictures and quotes that kept him building and trying through a lot of failures before anything worked. It's a pick-me-up for anyone shipping side projects when things get hard, and he adds new images regularly.

Unsloth's GGUF quantizations let you run DeepSeek-V4.1-Flash-1017 locally with llama.cpp or Unsloth Desktop. The model is MIT-licensed and has a 1M-token context. The smallest 1-bit quant needs a desktop with at least 128 GB of RAM (GPU VRAM counts toward it) and about 235 GB free on a fast NVMe SSD. A GPU is optional, and 64 GB of RAM works too, only slower.

A fan project runs the PC version of Need for Speed: Underground 2 (2004) in Chrome and Firefox. It's built on NFSU2Recomp, a static recompiler that translates the game's x86 code to C. On top of that it adds WebGPU graphics, Web Audio, up to 120 fps in widescreen, support for controllers, wheels and touch, free roam and online races with friends. It isn't affiliated with Electronic Arts.

Fernando Borretti argues that thinking, coding and research only last inside a community of other people, which gives you shared work to build on and the motivation of being useful to others. AI makes contributions like blog posts, open-source libraries and new programming languages superfluous, so those communities dry up. His examples are software engineering and now mathematics, and he expects individual intellectual work to fade with them.

The Residential Proxy Observatory detected residential proxy traffic exiting the White House network (AS2650) 236 times over 30 days. The traffic passed through 28 different proxy services.

Login code screens split a six-digit code into six boxes held together by JavaScript, which breaks paste, Backspace, SMS and email autofill, password managers and screen readers. The article shows these failures in live demos, including in react-otp-input, which hardcodes autocomplete off, and input-otp, which needs 767 lines to hide a single input. The fix is one plain input with autocomplete="one-time-code", inputmode="numeric", maxlength and a pattern, and no script.

Delta Inference Engine, a Windows port of the Linux engine Xolotl, runs all 552B parameters of DeepSeek-V4.1-Flash-1017 with 2-bit experts on a single consumer PC. It serves an OpenAI- and Anthropic-compatible API with up to 1M tokens of context. It needs an NVIDIA RTX card with 12 GB+ VRAM, 128 GB+ RAM and about 385 GB on a fast NVMe SSD. On an RTX 3090 with 128 GB RAM it writes about 33 tokens per second.

REA is an open-source MCP server and CLI that lets AI coding agents like Claude Code, Codex and Cursor inspect apps without their source code. It works with native binaries through Hopper, Ghidra or IDA, and with JavaScript and Electron apps, .NET assemblies, Android apps and websites. The agent can then explain how a feature works, show the evidence behind each conclusion and build a similar version for your project. Analysis runs locally, and setup is a single npx command.

North Korean operatives used a fake job interview to carry out a $6.3 billion Binance heist, according to the exchange. Binance said the incident was not a single isolated breach or a simple phishing attack but a broader, coordinated operation.

AI is reshaping cybersecurity by accelerating both attackers and defenders, exposing long-neglected technical debt in fragile legacy systems. The pressure is forcing organizations to confront and pay down decades of accumulated infrastructure problems.

Turkish software engineer Bedirhan Yildirim used AI coding agents and free Polish public data to build an explorable 3D replica of Warsaw in two days. The free online map has almost 200,000 buildings, trams that follow real timetables, live weather and 45,000 streetlamps that light up at night, plus a flight simulator. He says AI did 90% of the job, and the last 10% of human work keeps it from being "AI slop".

The ArtCraft team is building sixteen free, open-source desktop apps from scratch in Rust, with familiar layouts and shortcuts. They cover image editing, vector illustration, video, raw photos, PDFs, motion graphics, page layout, interface design, 3D texture painting, word processing, audio, slides, CAD and spreadsheets. They run natively on macOS, Windows and Linux without Electron, most also run in the browser through WebAssembly, and every app can be driven by AI agents through a CLI or an MCP server. All of them are still in early alpha or in development.

Rocky Linux founder Gregory Kurtzer had a single OpenAI Codex task (GPT-5.6 Sol) write Ginux in eleven days. The result is about 139,000 lines of Rust and test code for an x86-64 kernel that boots from UEFI without Linux and runs unmodified Linux programs, including Rocky Linux 10 with systemd. It's an experimental research project tested in QEMU, not a replacement for Linux, and it comes with a technical report on how he directed the agent.

C++26 was unexpectedly disapproved in an ISO National Bodies vote on October 23, 2026, despite being expected to pass as a routine formality. The decision casts uncertainty over the next C++ standard’s approval timeline.

Kitty 0.50 has been released, adding docked windows to the fast, cross-platform, GPU-based terminal. Users can update by following the project’s instructions.

Progress has been shared on a powerful cryptographic proof called a zkSNARK, which is uniquely suited to provable inference. The work advances cryptographic verification for large language model inference.

Tencent’s Hy4 is a strong open-weight model that brings open models back to the cost/accuracy frontier, with a notable game-development focus including adaptations for Unreal, Unity, and Godot and gains on GameDev and GameLogic benchmarks, while acknowledging gaps to top proprietary models in research math, long-horizon coding, and advanced visual QA. Tencent is also open-sourcing HyCell for single-cell representation and HyPDE for physical field modeling. In data analytics tests Hy4 is competitive and cheap but overly thorough and slow, running many queries per question.