Posted by david927 19 hours ago
Ask HN: What are you working on? (September 2026)
We created it as an alternative to the price gouging that happened over the last year at various timesheet SaaS companies. The idea is that good software should not cost 10x as much just because 10x as many people use it. In our opinion, software should also be functionally complete. There should be no arbitrary tiers that only exist to force people into more expensive plans.
The stack is Elixir with Phoenix LiveView, deployed on a Talos Kubernetes cluster. This is different from what we normally build on, and honestly quite fun.
Lately we have been thinking hard about the temporal nature of timesheet software. Many things get quite complicated once you factor in historical versions: the need to retire, lock, change, reopen records or rates, and so on. We have been working hard to polish Abejora to make everything accessible, user friendly, and intuitive.
The idea is simple: store state you'd normally store in a separate database in the document itself.
Then, also give the the document a bunch of powerful APIs granted to it by its host:
(1) Sync changes to the document with live collaborators
(2) Extract data from the document using a simple JSON map config
(3) Boot up a full content CMS from the same simple JSON map config
(4) More features added every month
I'm really excited with even the tiny bit of early traction I've gotten. About a dozen paying users so far.
Long term, I'd love to have this file type (vanilla HTML file + some extra features granted by the host) become a file type other companies offer tools and their own hosting for (like Wordpress).
Hosted platform: https://hyperclay.com
Native offline apps (source available, almost MIT license):
I'm a Senior Eng on Wall Street, so I have my own personal experience with this problem, but i'd love to connect with other leads, release managers, etc to get their thoughts, suggestions, and what their current process looks like.
It's just a fun game but can be useful for people wanting to get a better sense of what the real estate market is like at the moment.
I should really have added a bunch more games by now but I've been sidetrack. Mentioning it here has already given me a bit of motivation to do that now.
Any feedback will be much appreciated.
A programming language which lets you code multiplayer games like singleplayer games. Every program written in Easel is guaranteed deterministic and snapshottable, which is how it can automatically make your game multiplayer automatically using rollback netcode.
My hope is that teenagers who makes games in Scratch might like to make games in Easel because it means they can make games they can play with their friends, without having to deploy a server or learn anything about networking on synchronization. People can literally make a multiplayer game on their first day of coding.
Currently been working on adding tilemaps.
And then even for simple turn-based games, there has to be elaborate fully-decided like state diagrams for all the various types of disconnect and latency states so that gameplay doesn't desynch or hitch in awkward ways.
Been working on Feedbun for a little while, Elixir/Phoenix stack, with bespoke LLM fleets for production workflows. Recently just started using React Native (Expo) for the companion mobile app so I can use it when I shop IRL.
Also started working on a 2D strategy game using Godot engine, still super early days. Been fun to experiment with different LLMs capabilities on game development.
- Puma, The terminal browser: https://github.com/meerita/puma-browser
- Kernq, a full replacement of lsof: https://github.com/meerita/kernq
- Entroq, modern lossless compression for modern systems: https://github.com/meerita/entroq
Why? There's no why. The world doesn't need another Markdown editor. I still made one.
Everything is on-device, purposefully human-centric without being anti-AI. I built this because I wanted to work on a project that I could sweat the details unconstrained by deadlines, investors, etc.
I was frustrated with other writing experiences, and like most I gravitated to Apple Notes due to simplicity, but I still missed so many things. As someone that's spent the last 2 years immersed in Cursor, Codex and CC, I wished some of their input behaviors existed in a regular text editor.
For example, you can use / to add styles, or : to add emojis, shortcuts like < 3 for a heart, - > for an arrow, and so on. It honestly feels great.
Is it radically different than other editors? I don't think so. But I think it's well executed across every dimension possible.
Maybe one different thing: multiplayer. It's live editing like google docs, but P2P using WebRTC and strong cryptography. The limitation here is that you can only share a file while having it open yourself.
Great care was taken to performance and portability. This is one of the areas where I spent a ridiculous amount of time simply because I enjoyed pushing the limits of the stack. While using you will notice that rendering, GPU, RAM and bundle sizes were heavily optimized. The Mac app is < 50mb.
Very soon, I'll launch an iOS version with iCloud sync. In the future, I might launch basic AI features. If you miss a great text editor in your life, give it a try.
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.
Would be good to keep track of how long it takes you to complete, so you can compare it with how long it takes Opus and then Fable to accomplish the same thing.
I should have. All I have right now is a free Cursor account and Github Copilot. I've got to start spending money on tokens and get up to date. That article today by Carmack chewing out game devs for hand coding was striking, considering who he is and how much low level stuff he has done.
I'm doing 3D rendering in Rust. (Think three.js, but in Rust.) This is well outside the mainstream, so there's not much on the web to populate training sets. A year ago, AI coding tools could not cope with this area at all. Today, maybe.
I have no idea how good Claude is at understanding breaking changes to an API. I expect it would get simple changes, such as structure field renamings, without any trouble. But the gyrations required to handle write only access to GPU memory in Rust might not appear in the training data.