Thursday, 26 February 2026
Wednesday, 25 February 2026
New top story on Hacker News: Show HN: A real-time strategy game that AI agents can play
Show HN: A real-time strategy game that AI agents can play
23 by __cayenne__ | 3 comments on Hacker News.
I've liked all the projects that put LLMs into game environments. It's been a weird juxtaposition, though: frontier LLMs can one-shot full coding projects, and those same models struggle to get out of Pokémon Red's Mt. Moon. Because of this, I wanted to create a game environment that put this generation of frontier LLMs' top skill, coding, on full display. Ten years ago, a team released a game called Screeps. It was described as an "MMO RTS sandbox for programmers." The Screeps paradigm of writing code and having it executed in a real-time game environment is well suited to LLMs. Drawing on a version of the Screeps open source API, LLM Skirmish pits LLMs head-to-head in a series of 1v1 real-time strategy games. In my testing I found that Claude Opus 4.5 was the most dominant model, but it showed weakness in round 1 as it was overly focused on its in-game economy. Meanwhile, I probably spent a third of all code on sandbox hardening because GPT 5.2 kept trying to cheat by pre-reading its opponent's strategies. If there's interest, I'm planning on doing a round of testing with the latest generation of LLMs (Claude 4.6 Opus, GPT 5.3 Codex, etc.). You can run local matches via CLI. I'm running a hosted match runner with Google Cloud Run that uses isolated-vm. The match playback visualizer is statically served from Cloudflare. I've created a community ladder that you can submit strategies to via CLI, no auth required. I've found that the CLI plus the skill.md that's available has been enough for AI agents to immediately get started. Website: https://llmskirmish.com API docs: https://ift.tt/t0jUwg7 GitHub: https://ift.tt/47FLjnm A video of a match: https://www.youtube.com/watch?v=lnBPaZ1qamM
23 by __cayenne__ | 3 comments on Hacker News.
I've liked all the projects that put LLMs into game environments. It's been a weird juxtaposition, though: frontier LLMs can one-shot full coding projects, and those same models struggle to get out of Pokémon Red's Mt. Moon. Because of this, I wanted to create a game environment that put this generation of frontier LLMs' top skill, coding, on full display. Ten years ago, a team released a game called Screeps. It was described as an "MMO RTS sandbox for programmers." The Screeps paradigm of writing code and having it executed in a real-time game environment is well suited to LLMs. Drawing on a version of the Screeps open source API, LLM Skirmish pits LLMs head-to-head in a series of 1v1 real-time strategy games. In my testing I found that Claude Opus 4.5 was the most dominant model, but it showed weakness in round 1 as it was overly focused on its in-game economy. Meanwhile, I probably spent a third of all code on sandbox hardening because GPT 5.2 kept trying to cheat by pre-reading its opponent's strategies. If there's interest, I'm planning on doing a round of testing with the latest generation of LLMs (Claude 4.6 Opus, GPT 5.3 Codex, etc.). You can run local matches via CLI. I'm running a hosted match runner with Google Cloud Run that uses isolated-vm. The match playback visualizer is statically served from Cloudflare. I've created a community ladder that you can submit strategies to via CLI, no auth required. I've found that the CLI plus the skill.md that's available has been enough for AI agents to immediately get started. Website: https://llmskirmish.com API docs: https://ift.tt/t0jUwg7 GitHub: https://ift.tt/47FLjnm A video of a match: https://www.youtube.com/watch?v=lnBPaZ1qamM
Tuesday, 24 February 2026
New top story on Hacker News: The Missing Semester of Your CS Education – Revised for 2026
The Missing Semester of Your CS Education – Revised for 2026
40 by anishathalye | 1 comments on Hacker News.
We returned to MIT last month to teach a revised version of Missing Semester, six years after the original debut (which has been extensively discussed on HN, in https://ift.tt/UZcCP7p and https://ift.tt/Fp9YC7X ). We’ve updated the course based on our personal experiences as well as major changes in the field (e.g., the proliferation of AI-powered developer tools) over the past several years. The 2026 course includes revised versions of four lectures from the previous course, and it adds five entirely new lectures: - Development Environment and Tools - Packaging and Shipping Code - Agentic Coding - Beyond the Code (soft skills) - Code Quality We’d love to hear any feedback from the HN community to improve the current or future iterations of the course. In particular, we’re curious to hear the community’s take on our inclusion of AI-related topics (e.g., dedicating an entire class to the topic of agentic coding; though we tried to counterbalance it with plenty of disclaimers, and a dedicated section on AI etiquette in Beyond the Code). --Anish, Jon, and Jose
40 by anishathalye | 1 comments on Hacker News.
We returned to MIT last month to teach a revised version of Missing Semester, six years after the original debut (which has been extensively discussed on HN, in https://ift.tt/UZcCP7p and https://ift.tt/Fp9YC7X ). We’ve updated the course based on our personal experiences as well as major changes in the field (e.g., the proliferation of AI-powered developer tools) over the past several years. The 2026 course includes revised versions of four lectures from the previous course, and it adds five entirely new lectures: - Development Environment and Tools - Packaging and Shipping Code - Agentic Coding - Beyond the Code (soft skills) - Code Quality We’d love to hear any feedback from the HN community to improve the current or future iterations of the course. In particular, we’re curious to hear the community’s take on our inclusion of AI-related topics (e.g., dedicating an entire class to the topic of agentic coding; though we tried to counterbalance it with plenty of disclaimers, and a dedicated section on AI etiquette in Beyond the Code). --Anish, Jon, and Jose
New top story on Hacker News: AI-generated replies really are a scourge these days
AI-generated replies really are a scourge these days
17 by da_grift_shift | 17 comments on Hacker News.
17 by da_grift_shift | 17 comments on Hacker News.
Monday, 23 February 2026
Sunday, 22 February 2026
New top story on Hacker News: Show HN: Elecxzy – A lightweight, Lisp-free Emacs-like editor in Electron
Show HN: Elecxzy – A lightweight, Lisp-free Emacs-like editor in Electron
4 by kurouna | 1 comments on Hacker News.
Hi HN. I am a programmer from Japan who loves Emacs. I am building elecxzy. It is a free (zero-cost), lightweight, Emacs-like text editor for Windows. I designed it to be comfortable and ready to use immediately, without a custom init.el. Here is a quick overview: - Provides mouse-free operation and classic Emacs keybindings for essential tasks (file I/O, search, split windows, syntax highlighting). - Drops the Lisp execution engine entirely. This keeps startup and operation lightweight. - Solves CJK (Chinese, Japanese, Korean) IME control issues natively on Windows. I never managed to learn Lisp. I just copy-pasted snippets to maintain my init.el. However, I loved the Emacs keybindings. I loved operating an editor entirely without a mouse. I wanted an editor I could just open and use immediately. Also, standard Emacs binaries for Windows often have subtle usability issues for CJK users. So, I thought about whether I could build an Emacs-like text editor using Electron, the same framework as VS Code. Building an editor inside a browser engine required thinking a lot about what NOT to build. To make it feel native, I had to navigate DOM limitations. I learned that intentionally dropping complex features improves rendering speed. For example, I skipped implementing "word wrap." For syntax highlighting, I did not use a full AST parser. Instead, I used strict "line-by-line" parsing. The highlight colors for multi-line comments are occasionally incorrect, but it is practically unproblematic and keeps the editor fast. Under the hood, to bypass browser limitations and handle large files smoothly, I implemented a virtual rendering (virtual scrolling) system. For text management and Undo/Redo, I use a custom Piece Table. I built a custom KeyResolver for Emacs chords. I also used koffi to call Win32 APIs directly for precise IME control. I respect Windows Notepad as one of the most widely used text editors. However, in my daily work or coding tasks, I often felt it lacked certain features. On the other hand, I found VS Code too heavy just to write a quick memo. Even with extensions, it never quite gave me that native Emacs flow. I do not want to replace Notepad, VS Code, or Emacs. If users want rich extensions and heavy customization, I believe they should use Emacs or VS Code. My goal is to fill the gap between them—to build a "greatest common denominator" editor for people who just want an Emacs-like environment on Windows without the setup. It is still in alpha (so it might not work perfectly), but you can test it on Windows by downloading the zip from the GitHub releases, extracting it, and running elecxzy.exe. For screenshots, basic usage, and keybindings, please check the README on the GitHub project page. I am looking for feedback: Is there a demand for a zero-config, Lisp-free, "Notepad-like" Emacs-style editor? What are the minimum standard features required to make it useful? I would love to hear your technical insights.
4 by kurouna | 1 comments on Hacker News.
Hi HN. I am a programmer from Japan who loves Emacs. I am building elecxzy. It is a free (zero-cost), lightweight, Emacs-like text editor for Windows. I designed it to be comfortable and ready to use immediately, without a custom init.el. Here is a quick overview: - Provides mouse-free operation and classic Emacs keybindings for essential tasks (file I/O, search, split windows, syntax highlighting). - Drops the Lisp execution engine entirely. This keeps startup and operation lightweight. - Solves CJK (Chinese, Japanese, Korean) IME control issues natively on Windows. I never managed to learn Lisp. I just copy-pasted snippets to maintain my init.el. However, I loved the Emacs keybindings. I loved operating an editor entirely without a mouse. I wanted an editor I could just open and use immediately. Also, standard Emacs binaries for Windows often have subtle usability issues for CJK users. So, I thought about whether I could build an Emacs-like text editor using Electron, the same framework as VS Code. Building an editor inside a browser engine required thinking a lot about what NOT to build. To make it feel native, I had to navigate DOM limitations. I learned that intentionally dropping complex features improves rendering speed. For example, I skipped implementing "word wrap." For syntax highlighting, I did not use a full AST parser. Instead, I used strict "line-by-line" parsing. The highlight colors for multi-line comments are occasionally incorrect, but it is practically unproblematic and keeps the editor fast. Under the hood, to bypass browser limitations and handle large files smoothly, I implemented a virtual rendering (virtual scrolling) system. For text management and Undo/Redo, I use a custom Piece Table. I built a custom KeyResolver for Emacs chords. I also used koffi to call Win32 APIs directly for precise IME control. I respect Windows Notepad as one of the most widely used text editors. However, in my daily work or coding tasks, I often felt it lacked certain features. On the other hand, I found VS Code too heavy just to write a quick memo. Even with extensions, it never quite gave me that native Emacs flow. I do not want to replace Notepad, VS Code, or Emacs. If users want rich extensions and heavy customization, I believe they should use Emacs or VS Code. My goal is to fill the gap between them—to build a "greatest common denominator" editor for people who just want an Emacs-like environment on Windows without the setup. It is still in alpha (so it might not work perfectly), but you can test it on Windows by downloading the zip from the GitHub releases, extracting it, and running elecxzy.exe. For screenshots, basic usage, and keybindings, please check the README on the GitHub project page. I am looking for feedback: Is there a demand for a zero-config, Lisp-free, "Notepad-like" Emacs-style editor? What are the minimum standard features required to make it useful? I would love to hear your technical insights.
Saturday, 21 February 2026
Friday, 20 February 2026
Thursday, 19 February 2026
New top story on Hacker News: Show HN: A physically-based GPU ray tracer written in Julia
Show HN: A physically-based GPU ray tracer written in Julia
7 by simondanisch | 1 comments on Hacker News.
We ported pbrt-v4 to Julia and built it into a Makie backend. Any Makie plot can now be rendered with physically-based path tracing. Julia compiles user-defined physics directly into GPU kernels, so anyone can extend the ray tracer with new materials and media - a black hole with gravitational lensing is ~200 lines of Julia. Runs on AMD, NVIDIA, and CPU via KernelAbstractions.jl, with Metal coming soon. Demo scenes: github.com/SimonDanisch/RayDemo
7 by simondanisch | 1 comments on Hacker News.
We ported pbrt-v4 to Julia and built it into a Makie backend. Any Makie plot can now be rendered with physically-based path tracing. Julia compiles user-defined physics directly into GPU kernels, so anyone can extend the ray tracer with new materials and media - a black hole with gravitational lensing is ~200 lines of Julia. Runs on AMD, NVIDIA, and CPU via KernelAbstractions.jl, with Metal coming soon. Demo scenes: github.com/SimonDanisch/RayDemo
Wednesday, 18 February 2026
Tuesday, 17 February 2026
Monday, 16 February 2026
Sunday, 15 February 2026
Saturday, 14 February 2026
Friday, 13 February 2026
Thursday, 12 February 2026
New top story on Hacker News: Show HN: A free online British accent generator for instant voice conversion
Show HN: A free online British accent generator for instant voice conversion
4 by Katherine603 | 9 comments on Hacker News.
I've developed a simple AI-powered British accent generator. Enter or paste your text, select the voice that best fits your project's tone, and generate speech for free. It supports up to 500 characters and offers 8 distinct, lifelike voices. Everything runs entirely within your browser. I'm primarily seeking feedback on output quality, user experience, and any technical improvements worth exploring.
4 by Katherine603 | 9 comments on Hacker News.
I've developed a simple AI-powered British accent generator. Enter or paste your text, select the voice that best fits your project's tone, and generate speech for free. It supports up to 500 characters and offers 8 distinct, lifelike voices. Everything runs entirely within your browser. I'm primarily seeking feedback on output quality, user experience, and any technical improvements worth exploring.
Wednesday, 11 February 2026
New top story on Hacker News: Show HN: Itsyhome – Control HomeKit from your Mac menu bar (open source)
Show HN: Itsyhome – Control HomeKit from your Mac menu bar (open source)
15 by nixus76 | 8 comments on Hacker News.
Hey HN! Nick here – developer of Itsyhome, a menu bar app for macOS that gives you control over your whole HomeKit fleet (and very soon Home Assistant). I run 130+ HomeKit devices at home and the Home app was too heavy for quick adjustments. Full HomeKit support, favourites, hidden items, device groups, pinning of rooms/accessories/groups as separate menu bar items, iCloud sync – all in a native experience and tiny package. Open source ( https://ift.tt/zpbseV1 ) and free to use (there is an optional one-time purchase for a Pro version which includes cameras and automation features). Itsyhome is a Mac Catalyst app because HomeKit requires the iOS SDK, so it runs a headless Catalyst process for HomeKit (and now Home Assistant) access while using a native AppKit plugin over a bridge protocol to provide the actual menu bar UI – since AppKit gives you the real macOS menu bar experience that Catalyst alone can't. It comes with deeplink support, a webhook server, a CLI tool (golang, all open source), a Stream Deck plugin (open source, all accessories supported), and the recent update also includes an SSE event stream (HomeKit and HA) - you can curl -N localhost:8423/events and get a real-time JSON stream of every device state change in your home. Home Assistant version is still in beta – would anyone be willing to test it via TestFlight? Appreciate any feedback and happy to answer any questions.
15 by nixus76 | 8 comments on Hacker News.
Hey HN! Nick here – developer of Itsyhome, a menu bar app for macOS that gives you control over your whole HomeKit fleet (and very soon Home Assistant). I run 130+ HomeKit devices at home and the Home app was too heavy for quick adjustments. Full HomeKit support, favourites, hidden items, device groups, pinning of rooms/accessories/groups as separate menu bar items, iCloud sync – all in a native experience and tiny package. Open source ( https://ift.tt/zpbseV1 ) and free to use (there is an optional one-time purchase for a Pro version which includes cameras and automation features). Itsyhome is a Mac Catalyst app because HomeKit requires the iOS SDK, so it runs a headless Catalyst process for HomeKit (and now Home Assistant) access while using a native AppKit plugin over a bridge protocol to provide the actual menu bar UI – since AppKit gives you the real macOS menu bar experience that Catalyst alone can't. It comes with deeplink support, a webhook server, a CLI tool (golang, all open source), a Stream Deck plugin (open source, all accessories supported), and the recent update also includes an SSE event stream (HomeKit and HA) - you can curl -N localhost:8423/events and get a real-time JSON stream of every device state change in your home. Home Assistant version is still in beta – would anyone be willing to test it via TestFlight? Appreciate any feedback and happy to answer any questions.
Tuesday, 10 February 2026
New top story on Hacker News: AI doesn’t reduce work, it intensifies it
AI doesn’t reduce work, it intensifies it
36 by walterbell | 8 comments on Hacker News.
Original article: https://ift.tt/13rMT8K...
36 by walterbell | 8 comments on Hacker News.
Original article: https://ift.tt/13rMT8K...
Monday, 9 February 2026
New top story on Hacker News: Show HN: Minimal NIST/OWASP-compliant auth implementation for Cloudflare Workers
Show HN: Minimal NIST/OWASP-compliant auth implementation for Cloudflare Workers
11 by vhsdev | 4 comments on Hacker News.
This is an educational reference implementation showing how to build reasonably secure, standards-compliant authentication from first principles on Cloudflare Workers. Stack: Hono, Turso (libSQL), PBKDF2-SHA384 + normalization + common-password checks, JWT access + refresh tokens with revocation support, HTTP-only SameSite cookies, device tracking. It's deliberately minimal — no OAuth, no passkeys, no magic links, no rate limiting — because the goal is clarity and auditability. I wrote it mainly to deeply understand edge-runtime auth constraints and to have a clean Apache-2.0 example that follows NIST SP 800-63B / SP 800-132 and OWASP guidance. For production I'd almost always reach for Better Auth instead ( https://ift.tt/DoKkNQp ) — this repo is not trying to compete with it. Live demo: https://private-landing.vhsdev.workers.dev/ Repo: https://ift.tt/filwPa9 Happy to answer questions about the crypto choices, the refresh token revocation pattern, Turso schema, constant-time comparison, unicode pitfalls, etc.
11 by vhsdev | 4 comments on Hacker News.
This is an educational reference implementation showing how to build reasonably secure, standards-compliant authentication from first principles on Cloudflare Workers. Stack: Hono, Turso (libSQL), PBKDF2-SHA384 + normalization + common-password checks, JWT access + refresh tokens with revocation support, HTTP-only SameSite cookies, device tracking. It's deliberately minimal — no OAuth, no passkeys, no magic links, no rate limiting — because the goal is clarity and auditability. I wrote it mainly to deeply understand edge-runtime auth constraints and to have a clean Apache-2.0 example that follows NIST SP 800-63B / SP 800-132 and OWASP guidance. For production I'd almost always reach for Better Auth instead ( https://ift.tt/DoKkNQp ) — this repo is not trying to compete with it. Live demo: https://private-landing.vhsdev.workers.dev/ Repo: https://ift.tt/filwPa9 Happy to answer questions about the crypto choices, the refresh token revocation pattern, Turso schema, constant-time comparison, unicode pitfalls, etc.
New top story on Hacker News: Show HN: Algorithmically Finding the Longest Line of Sight on Earth
Show HN: Algorithmically Finding the Longest Line of Sight on Earth
16 by tombh | 5 comments on Hacker News.
We're Tom and Ryan and we teamed up to build an algorithm with Rust and SIMD to exhaustively search for the longest line of sight on the planet. We can confirm that a previously speculated view between Pik Dankova in Kyrgyzstan and the Hindu Kush in China is indeed the longest, at 530km. We go into all the details at https://alltheviews.world And there's an interactive map with over 1 billion longest lines, covering the whole world at https://map.alltheviews.world Just click on any point and it'll load its longest line of sight. Some of you may remember Tom's post[1] from a few months ago about how to efficiently pack visibility tiles for computing the entire planet. Well now it's done. The compute run itself took 100s of AMD Turin cores, 100s of GBs of RAM, a few TBs of disk and 2 days of constant runtime on multiple machines. If you are interested in the technical details, Ryan and I have written extensively about the algorithm and pipeline that got us here: * Tom's blog post: https://ift.tt/Rfxgu0l * Ryan's technical breakdown: https://ift.tt/KVE5LPR This was a labor of love and we hope it inspires you both technically and naturally, to get you out seeing some of these vast views for yourselves! 1. https://ift.tt/anFKmEb
16 by tombh | 5 comments on Hacker News.
We're Tom and Ryan and we teamed up to build an algorithm with Rust and SIMD to exhaustively search for the longest line of sight on the planet. We can confirm that a previously speculated view between Pik Dankova in Kyrgyzstan and the Hindu Kush in China is indeed the longest, at 530km. We go into all the details at https://alltheviews.world And there's an interactive map with over 1 billion longest lines, covering the whole world at https://map.alltheviews.world Just click on any point and it'll load its longest line of sight. Some of you may remember Tom's post[1] from a few months ago about how to efficiently pack visibility tiles for computing the entire planet. Well now it's done. The compute run itself took 100s of AMD Turin cores, 100s of GBs of RAM, a few TBs of disk and 2 days of constant runtime on multiple machines. If you are interested in the technical details, Ryan and I have written extensively about the algorithm and pipeline that got us here: * Tom's blog post: https://ift.tt/Rfxgu0l * Ryan's technical breakdown: https://ift.tt/KVE5LPR This was a labor of love and we hope it inspires you both technically and naturally, to get you out seeing some of these vast views for yourselves! 1. https://ift.tt/anFKmEb
Sunday, 8 February 2026
Saturday, 7 February 2026
Friday, 6 February 2026
Thursday, 5 February 2026
Wednesday, 4 February 2026
Tuesday, 3 February 2026
Monday, 2 February 2026
Sunday, 1 February 2026
Subscribe to:
Comments (Atom)