Live data from Hacker News

Ask HN: What Are You Working On? (April 2026)

news.ycombinator.com

651–660 of 1001 posts

Re: Ask HN: What Are You Working On? (April 2026)

#651

I'm building Tela ( https://github.com/paulmooreparks/tela ), a self-hosted relay that tunnels TCP services through encrypted WireGuard connections. The key difference from Tailscale and similar tools is that it requires no TUN adapter, no root access, and no admin privileges on either end. It runs entirely in userspace. My initial motivation was wanting to RDP and SSH into my home workstation from a locked-down corp…

I've been working on a similar tool for a while - connet (https://github.com/connet-dev/connet). It builds upon QUIC (instead of wireguard), but I think from an enduser perspective the results ends up looking pretty similar.

Re: Ask HN: What Are You Working On? (April 2026)

#653
I'm building a chaotic, physics-driven first-person woodcutting game: https://store.steampowered.com/app/4521770/Drunk_Woodcutter/

I was splitting wood in real life, and even though it's a chore, it's oddly satisfying work, too. Every hit feels different, sounds different, every piece of wood splits in different ways. I want to catch that vibe and put a good dose of comedy on top.

It's using realtime mesh cutting for the splitting, and I recorded 48 different chopping sounds for it :D

Re: Ask HN: What Are You Working On? (April 2026)

#654
Me and a buddy of mine were sick of all things AI in our daily jobs, so we started to play around with different ways to use LLMs and landed on PitchSlap (https://pitchslap.gg).

You create arenas, which are thematic problem statements, submit ideas to solve it, and then vote on the best one, all with help of your avatar. The idea is to use AI as a part of the ideation process, which could be for things like hackathons (what can we build to solve x), evaluating business ideas, or in general just mess around with models. And the whole thing is wrapped in a high-concept anime corporate parody style.

There are also battles, which are shorter simpler versions of arenas and showdowns, which are supposed to be a follow up to arena, where you flesh out the winning idea - answer a set of questions about the idea and again vote on the best answer.

We have a bunch of features and ideas to take the concept forward (credit style economy, where you play for "slaps"; fully automated mode, where AI driven avatars hash out the whole game-play cycle and you can observe, RPG-style attributes for the avatars that have more significant impact to the pitch generation process), but have in general enjoyed the process of using these technologies for something a little less serious.

Re: Ask HN: What Are You Working On? (April 2026)

#658
Not a big thing, but I’m happy to work again on a side project I started last year and abandoned because of some limitations from the stack I chose.

It’s https://napodico.it, a very simple online Neapolitan dictionary. Nothing extraordinary here but it’s something that will tremendously help me organize my notes on Neapolitan words and expressions that are currently scattered across Google docs. I’ve built https://www.schedarionapoletano.it in the past but it’s from a dictionary made my someone else and I didn’t want to mix my own definitions in.

This is not a 'product', it’s a normal free website as we used to do before the commercial Web became the norm. The front is very simple, but the entries can be edited in the back a bit like a CMS but with Wikipedia-like inter-links and redirects.

The stack I started with was SvelteKit and Drizzle for the ORM, but I quickly hit several limitations of Drizzle and I abandonned the project. Last week I asked Claude to split it into the more familiar stack that I used to use before trying to fit everything into a single TS project that is: one Python app (FastAPI/SQLAlchemy) for the backend that exposes a REST API consumed by the front app (SvelteKit). This is a lot more flexible, and it has allowed me to work again on the project.

Re: Ask HN: What Are You Working On? (April 2026)

#659

I'm working on a fully offline, client-side train journey planner for UK rail - https://railraptor.com When booking flights, I use sites like Kiwi and Skyscanner that let you do flexible searches - multiple destinations, custom connections, creative routes, etc. But rail search feels oddly constrained. All the UK train operators offer basically the same experience, and surface the exact same routes. I always suspecte…

Great that it stores the entire timetable in only 6MB(?) of storage.

Some feedback: I don't think it can route through London as it isn't aware of tube connections between stations? And the classic stress test of Penzance to Thurso is too long for the routing algorithm, but I imagine that's beyond scope?

Pricing would make this a super useful tool!

Post reply on HN