Ask HN: What are you working on? (May 2026)
12 comments
I just started on an open source and open weight supervised learning model to recognize japanese kanji characters drawn on the screen.
I have a working prototype written in Julia which is a very simple neural network. The input is in vector format so traditional convolutional neural networks don’t work out of the box but I swapped the convolution layer with a path simplification algorithm and it worked extremely well. Like 20 samples per character (from a set of only 5 hiragana during prototype phase) was enough to get 100% accuracy in a test collection of 5 samples per character after only 30 iterations of training.
I plan an working with free and open data, which I don‘t think exists for japanese kanji characters (at least not in vector format; KanjiVG only has one sample per character and I need dozens) so I also build a crowdsourcing web site to collect data from random people on the internet.
I am planning to run some more experiments with my prototype model before I release the crowdsourcing web page to an actual server though.
Model prototype: https://github.com/runarberg/kantoku-prototype
Crowdsource app: https://github.com/runarberg/kantoku-collector
I have a working prototype written in Julia which is a very simple neural network. The input is in vector format so traditional convolutional neural networks don’t work out of the box but I swapped the convolution layer with a path simplification algorithm and it worked extremely well. Like 20 samples per character (from a set of only 5 hiragana during prototype phase) was enough to get 100% accuracy in a test collection of 5 samples per character after only 30 iterations of training.
I plan an working with free and open data, which I don‘t think exists for japanese kanji characters (at least not in vector format; KanjiVG only has one sample per character and I need dozens) so I also build a crowdsourcing web site to collect data from random people on the internet.
I am planning to run some more experiments with my prototype model before I release the crowdsourcing web page to an actual server though.
Model prototype: https://github.com/runarberg/kantoku-prototype
Crowdsource app: https://github.com/runarberg/kantoku-collector
https://un-project.org/ --- A web app to explore United Nations meetings by browsing transcripts, speeches, and voting records from the General Assembly and Security Council.
It features some cool visualizations like this animated voting bubble chart: https://un-project.org/votes/bubble/
and this voting similarity map (click a country to see its closest matches): https://un-project.org/votes/map/
It features some cool visualizations like this animated voting bubble chart: https://un-project.org/votes/bubble/
and this voting similarity map (click a country to see its closest matches): https://un-project.org/votes/map/
These are typically posted at the end of the month. When people just rando post them every few days/week, all it does is create noise and dilute the meaning of everyone's post.
https://github.com/born-ml/born --- Production-ready ML framework for Go with zero dependencies. Train and deploy neural networks as single binaries. PyTorch-like API, type-safe tensors, automatic differentiation.
https://tickstem.dev — cron scheduling, uptime monitoring, heartbeat checks, and email verification under one API key. Go + Node.js SDKs, MCP server.
https://github.com/gogpu --- Pure Go GPU Computing Ecosystem — Graphics, Shaders, ML, GUI. Zero CGO.
----- I'm working on rebuilding this OSINT Repository website: https://osintguide.com/