YHK - yet another hacker news clone

I’m building a website for discovering first names that’s enjoyable to explore: https://abracadanames.com

It started with a family member telling me they were frustrated with existing websites. What I noticed is that in many cases they feel cluttered and not smooth to use, almost like a 1-to-1 translation of the underlying database queries, where they have you choose filters and sorting options upfront then browse the paginated results.

I wanted to avoid all of that by creating a different experience; I built an interactive full-screen wall UI with infinite browsing so that most of the screen space is dedicated to the core content. I worked on the name sampling algorithm to try to make it as enjoyable to browse as I could, and the only filter displayed by default is the one to select genders.

More advanced queries are available through the search field which supports natural language queries, running on a local multilingual interpretation engine with a cloud LLM fallback. The database engine is custom built with Rust, compiled to WASM for higher performance when running on CloudFlare Workers. I don’t use an external DB service for names today.

What I’m pretty proud of is to know a lot of technical work is hidden behind that UI, and how it has been shaped by the experience I wanted instead of the other way around.

I've been working on https://pronto.stream/, I wanted to mess around with MCP and building a world news system to power the decision making and surveillance of other systems I would likely curate.

I was curious about formats more efficient for agent communication than JSON and also proving if the framework I curated could handle the load and traffic.

It's been really fun to observe.

The 2026-27 soccer season just started.

If you love catching up on highlights, check https://thepelota.tv/

It’s like Netflix but for soccer highlights, with all the major leagues in one place as soon as they are available on YouTube, all free and no ads.

I've been working on a typed, functional-imperative, full stack programming language with IDE support: https://www.firefly-lang.org/

I'm looking for land to buy in Wyoming to build my AI apocalypse bunker on.

- danterm (https://github.com/danneu/danterm): a better terminal than iterm, ghostty, kitty, etc.

- braid (https://github.com/danneu/braid): a luks+btrfs tool for managing an encrypted disk pool on my NixOS NAS.

- tes3-es (https://github.com/danneu/tes3-es): a comprehensive latam translation of Morrowind.

I really like Morrowind's data file format -- how it's just a sequence of records -- except how bad it is for localization.

Basically, you provide a key for quest/info dialogue like "the ruins", and any time "the ruins" appears in dialogue, the text is hyperlinked such that if you click it, it activates the target dialogue.

But this implicit behavior means that even the production game has a lot of accidental hyperlinks and some bugs like orphaned dialogue branches because someone changed the text to "those ruins" not realizing "the ruins" was critical.

So I used a bunch of LLMs to build a glossary system and lint tools to check consistency. Ideally OpenMW would implement some QoLs here like a l10n dictionary system + markup like "I heard about {the ruins|1823432}."

I'm trying to build a tool site that has all the tools you'll ever need. I slowly add some each week.

https://allthedamn.tools

I’ve been hacking on a single Anki extension for… the last three years now:

https://smart-notes.xyz

The idea is to make an extension that allows you to flexibly fill out fields on your cards with AI generated text, speech, and images. Think generating example sentences for your entire deck, or adding furigana to kanji (small phonetic characters with a very delicate syntax). It works just as well for language learners as it does for medical students, etc.

The secondary goal has been to make this the most well designed, intuitive, and stable Anki extension in the world. I don’t think it’s there yet, but it’s on its way. It’s been very interesting to try to work out a design that affords the flexibility to generate anything for whatever you’re studying but remains approachable and discoverable to your marginal Anki user, many of whom are younger and have varying levels of English ability. Many interesting technical challenges as well.

Slowly building towards the Show HN post, but need a few more features. Connecting with and chatting with dedicated users in countries around the world over the last few years has made this the most gratifying thing I’ve ever built! It’s still at the scale that receiving support emails is fun and I can build relationships with power users and long term subscribers.

- i dropped a neural net jane street puzzle (https://huggingface.co/spaces/jane-street/droppedaneuralnet) very good mindbreaker, learned a lot of new things on the way

I've continued working on my game project, Vestiges, which is a 2D narrative strategy rogue-like in Godot. In brief: I once had a career in game development, in game design or project direction roles. Life circumstances derailed my career some years ago, but earlier this year a colleague encouraged me to experiment with genAI. I began Vestiges to see if I could create something again, to enjoy the creative process. It progressed well enough that I'm planning to complete and release it.

For various reasons unrelated to the project, progress since last month has been slow (and I expect the next month will be as well). Most of the work I've done has been for peripheral features, not anything on the critical path. After my brother bounced off of the narrative elements (not unexpected :) ), I added an "aliterate" lens that distills the game into its strategy gameplay, burying the narrative content for those who aren't interested in it.

In preparation for releasing a demo version, I added opt-in collection of a session's gameplay data, which I intend to analyze to find ways to improve the UI, UX, gameplay, balance. By default, no identifying information whatsoever is collected. A separate opt-in adds minimal technical information (things like OS, CPU, GPU, display settings).

I added global text size option as an accessibility feature and used an MCP server to have Claude do the initial layout testing. This was an interesting experiment, but requires enough tokens that it's not very practical as a frequent part of development. These features went hand-in-hand -- via the MCP server, Claude takes a screenshot every five seconds to decide what to hover over or click on next. These screenshots consume tokens rapidly, so they are reduced in size. With a larger font size, the screenshots can be reduced further with the text remaining legible to Claude.

Other housekeeping: numerous iterative improvements to my general development process and use of Claude. I wrote my first simple skill: used in conjunction with the brainstorming skill from the superpowers plugin, Claude creates an "origin" document that preserves my initial prompt and some aspects of the brainstorming discussion.

I expect to launch a demo version once my current life situation settles a bit. I'm still not in any particular hurry, but it would be good to start getting player feedback. Also, to the whatever extent I do want Vestiges to have any commercial success, the largest obstacle I'll face is discoverability and marketing.

A meta-comment: When I first mentioned this project, in the July "What are you working on?" (https://news.ycombinator.com/item?id=48886151), I explained some back story of this project. It got a fair amount of attention, rising to the top of the post and encouraging some discussion. Last month (https://news.ycombinator.com/item?id=49234405), I didn't repeat my story and received no upvotes or comments. This presents me with a quandary because it feels uncomfortable/self-indulgent to repeat my story each time, yet without that context, my project isn't particularly notable or interesting. (Or maybe luck/timing was a factor?)

https://grandpacad.com - AI modeling for 3D printing Been at it for about year and a half.

Really exciting stuff is happening literally every month, because underlying models are getting better and better. When I started it was pretty basic: “make a cube with a hole through it”. Now it’s at the point “make a raspberry pi 4 case” and the agent searches, builds, verifies…

What surprised me in this process is how little meaning AI benchmarks have. Pareto frontier for my use case looks completely different than any other benchmark portrays.

Build one of the best emoji sites there is.

https://emojistime.com

I'm decompiling an obscure 1998 computer game using an LLM agent skill that I'm simultaneously developing from papers and books on reverse-engineering (a subject I've never previously engaged with) with the aim of migrating it (well, at least its assets) to the browser: https://esoteria.pages.dev

Here's the skill: https://github.com/gavmor/x86-cpp-reversing-skill

It's all slop, but it's battle-tested and producing results. It's quite fun and inspiring to see some of these assets like textures and geometry that are improperly decompiled, bit shifted or something like that. produces a lot of abstract, glitchy art.

To make sure that my process and results are reproducible, I've gotten back into maintaining Concourse CI pipelines. They were very useful when we were shipping a fork of k8s for VMware, and now they are a fun way of shipping cyberpunk billboard sprites to Cloudflare buckets.

I got back into Concourse in order to attempt a more rigorous approach to generative AI experimentation as well. Unfortunately, there are just so many odd techniques, configurations, loras, and utilities striking my fancy that my experiment pipelines have exploded into the dozens, and Concourse is not a large enough or a high-level enough organizing principle.

Software to promote live music and drink specials so I don't have to check dozens of Instagram accounts to find out what's happening tonight.

https://nittanynights.com/

https://indyafterfive.com/

I wrote my own site parsing toolkit which removed the friction I have with the current ones. It's now easy to add new locations quickly to scale.

Slowly putting my photography online again at https://flaneurphoto.com - in the middle of composing soundscapes for some of the works.

Finally getting around to organising my photo archive and setting up a metadata extraction pipeline that makes sense for me.

(At the day job, curious about adapting app layouts to accommodate the iPhone Duo and address the various Android foldables at the same time, too).

Haven't added a demo to the repo yet (will do soon) but building this TUI [1] in Go to manage my agent skills.

I don't like putting 20-30 agent skills in the .claude/skills/ dir and letting the agent figure it out. I keep all the skills I've created and adapted over time in a separate git repo and then from there I copy them to the project skill dir when i need them for a session and then remove them when I'm done.

This TUI just replaces all the manual work with ls, cp -r, and rm -rf commands with a few keystrokes.

[1] https://github.com/primaprashant/sei

I've been working on a collection of native AI/ML operators for TouchDesigner on macOS. Think of it as a happy marriage between the interface of TouchDesigner but with some of the features you know from ComfyUI.

https://www.patreon.com/MickeyvanOlst/posts/meet-aml-macos-1...

Typically TD for the Mac has always been a bit underserved, by leaning into some of the frameworks that exclusively exist on macOS I'm hoping more people will see it as a viable platform for this kind of stuff.

I've been building a framework for web, ios and android native development at https://github.com/vmsp/flypath.

It's based on the Server-Driven UI philosophy where the backend streams both code and UI into a frontend that just knows how to render what it's told. This is pretty cool because it makes it makes over-the-air updates possible.

It also follows the batteries-included philosophy of Rails. Has an ORM, background and cron jobs (all based on Postgres), Django-style migrations, JSX Emails and pretty great FFI that's built around a simple `"use native"` directive, in the React Server Component fashion).

Trying to reduce my doomscrolling IG and FB usage

I created https://getnoloop.com to block the feed on iOS

On a mission to prove you don't need YC to be successful.

TagLib-Wasm, which is to TypeScript/JavaScript what Mutagen is to Python.

https://charleswiltgen.github.io/TagLib-Wasm/

It works in browsers, Node.js, Deno, Bun, Cloudflare Workers, and Electron apps.