Ask HN: What are you working on? (September 2026)

Updating a Rust renderer from WGPU 24 to WGPU 30. 640 lines of fixes needed so far, and more to do. I've fixed all the easy stuff, and now I'm down to reverse engineering breaking changes of excessive cleverness.

I wonder if Claude Code could do this. That's a useful metric for the future - if a coding agent can't handle the breaking changes in a package upgrade, it's the package's fault. The day may come when that's something Github Actions checks for pull requests.

One idea I had for getting to the tail: filter out sites that appear to use SSO.

I'm working on TableForge[0], it's a browser based, solo or multiplayer, D&D 5e game. In TableForge, the DM is agentic with access to tools strictly following 5e rules. The DM is responsible for narration and reacting to players but your character sheet, inventory, spells are all real server resources you manage. The DM can interact with them through deterministic 5e-based tools (dice rolls, damage, sheet updates, memory). Players can play in real time or async. You can provide the DM a premise (or pick one from the library) and it'll flesh out a full campaign story arc. Either way it's a fresh story arc reacting to your actual decisions, every time.

It’s been really fun working on it and especially rewarding seeing the game bring friends together.

[0] https://tableforge.gg/

Posted this before, but Odds Assist Pro (http://pro.oddsassist.com/) and Sportsbook API (https://sportsbookapi.com/)

The Odds Assist tool started out as a way for me to quickly QA the data from the API. It then turned into its own product as an odds scanner and sports book advantage finder (mainly arbitrage, +EV, middles, etc). Over the past few months it's really pivoted toward prediction markets and collecting and displaying data to get an edge on markets that are soft.

The pivot to focus on PMs has been pretty fun from a development perspective since there's so many niche markets to look explore. I've always liked math, I'm a bit of a digital hoarder, and I've got ADHD so finding random things to collect a bunch of data for just to test a hypothesis just hits my brain in all the right ways. It's also given me an excuse to learn more about ML what has been great.

I'm working on a p2p streaming/sharing app called bubbleBASED. Its here https://bubblebased.com/ its based! on bubbles! But it uses Webtorrent to stream and share content in context bubbles. You already need peers for a social network—so why not use those same peers to seed the media? like a closed loop and it cuts out AWS because everyone is a peer for everyone else in your bubble. There is an assist from Heroku and Pinata. There is no algo. Its just whatever you post. Also you can add an affiliate link (from cj.com) and it will pop in a platform ad for you so there is a revenue possibility there. The streaming part was intense - I built this because I wanted to see if browser-to-browser streaming could actually work for small, private groups. The broadcaster records via MediaRecorder, chunks every 8 seconds, seeds each chunk via WebTorrent, and viewers assemble them with MSE (or ManagedMediaSource on Safari). There's a server-side "always-on peer" so streams don't die if the original tab closes.

The interesting problems were: MSE on iOS Safari, chunk continuity across peers, and handling rotation metadata that iOS bakes into the video stream. Also, the tracker is a separate bittorrent-tracker instance because the built-in WebTorrent trackers are unreliable for sustained swarms. Also i used a ton of graphql, so normally with a torrent you have to kind of let people find you but i used a graphql subscription to TELL people what the magnet links were called so they can fetch them asap. Lots of shortcuts like that, its been interesting!!