Ask HN: What Are You Working On? (December 2025)
What is everyone working on?
Every week I pull all the new talk recordings from hundreds of conferences (Devoxx, KubeCon, PyCon, QCon, LeadDev, JSNation, and many more) and even more podcasts podcasts. I feature the ones I think are must-watch with short summaries written by me, then include a list of everything else uploaded that week.
It started as a personal project to fix my own messy YT subscriptions and RSS feeds and now 7,500+ people read it.
I also publish extra editions from time to time like “The Most Watched Talks of 2024” which made it to the HN front page.
If you watch software engineering conference talks or listen to podcasts, you might find it useful. I’d love to know what you think!
There is also a way to search for articles using vectors, it's called "Semantic Search". So basically you can ask, for example, "Postgresql and how to best optimize it." and it would search for articles touching that subject, or at least related to it.
Wondering about the best way I can add a weekly newsletter built on top of the content currently being ingested, and still looking for more sources to add to the database (let me know if you have any good recommendations).
Probably very few creators care one way or another, as the links are going to the original content. Just interested if people had an opinion on the matter.
I try to avoid altering the original content as much as possible. I do need to sanitize and adjust parts of it to produce clean text on my site, but I’m careful not to change anything in a way that misrepresents the source. Only a few short phrases appear on GreatReads, and users cannot read the full article without visiting the original source.
I use semantic search ocasionally when building extra editions and use a sentence transformer with all-MiniLM-L6-v2 model.
How did you go about implementing semantic search in your app?
Currently spending time establishing relationships with historical societies, as I really need them to contribute points of interest, and stories. Many of these societies are run on a voluntary basis by 70+ year olds, so it's a long process. Getting some good responses eventually though, so it might actually go somewhere, just a lot slower than I want.
Also still doing https://wheretodrink.beer, but haven't added anything of note since playing on this other project.
And react2shell was a fun time
Attracting new monthly sponsors and people willing to buy me the occasional pizza with my crappy HTML skills.
No real monetization plans yet - just experimenting / improving the detection at the moment.
I am currently working on fifth grade math. My plan is to cover first grade math up to Calculus and High School Physics. I envision it as a companion tool for Khan Academy/Math Class/Math Books. It is inspired by Chris McMullen's math workbooks.
Check out the demo. No signup required. Progress is only stored locally. https://demo.numerikos.com/
We aim to provide a fast JIT ssh cert attestation.
With focus on: * making on/offboard users fast * efficient workflows (no need to lookup passwords for logins or sudo) * mitigate private key leaks (especially in BYOD/BYOK environments) * Help admins manage server access fast
GitHub(WIP): https://github.com/flotte-sh
And thanks for checking us out :)
The first release will be a map of median rent per square meter. Simple stuff.
It's built on three principles: - Simple Listening: Short, anonymous weekly surveys that take less than 60 seconds for the team to complete. - Actionable Insights: AI analyzes comments to identify recurring themes, so you don't have to read hundreds of open-ended responses. - Performance-Focused: The goal isn't just to measure 'happiness', but to provide suggestions that help managers make better decisions.
I was inspired to build it after my own experience as a manager, where I realized I wasn't monitoring my team's well-being closely enough and it was starting to impact our performance.
It's still very early days, and I'm looking for a few first users to give honest, brutal feedback on the concept and the product.
Any feedback on the idea or the product is appreciated.
I wonder if you could do some automated diffing between the scenario results. If they are pretty similar and they were previously considered in a good state, then you don't necessarily need a QA person to review it.
You can't use this without having at least basic knowledge of the command-line.
Currently in closed beta, but sending out new batches of invites frequently.
The idea of turning this into a simulation game is a possibility hence "warz" in the name.
It is a bunch of Jupyter notebooks being served by Voila. The "Blank.ipynb" one is more interactive but the ergonomics need work.
It is running free tier hosting so it could swamped if enough people use it simultaneously:
https://sportwarzsim-production.up.railway.app/
https://github.com/daltontf/SportWarzSim
There are instructions to run locally via Docker in the README.md
I called it Wosp for word-oriented search and print. See the GitHub page for more information: https://github.com/atrettel/wosp
This allows library authors to do more, like defining webhook handlers and (simple) database operations. The idea is to move complexity from the user to the author, making integrations easier.
I think libraries being allowed to write to your database is a pretty powerful concept, and can enable a number of interesting use cases.
The goal is to treat LLMs as constrained components inside explicitly defined workflows: strict input/output schemas, validated DAGs, replayable execution, and observable failure modes. Most instability I’ve seen in production AI systems isn’t model-related — it comes from ambiguous structure around the model.
We’re exploring this through a project called FACET, focused on making AI behavior testable and debuggable rather than probabilistic and opaque.
Early days, but the direction is clear: less magic, more contracts.
My main personal project is a self-hosted Hotjar alternative: https://www.uxwizz.com
Micdrop is already used by other products to implement robust voice conversation with AI in webapps. Cheaper and more controllable than realtime GPT.
Feedbacks are appreciated!
Built using our full-stack library toolkit Fragno [0].
[0]: https://fragno.dev/
Local: https://github.com/basicmachines-co/basic-memory or hosted: http://basicmemory.com
You can fight monsters, craft, and conjure magic by playing real notes on your actual guitar. Web Browser Audio API handles pitch detection.
Basically, trying to make guitar practice a bit more fun by adding gamification.
Free demo: https://openfret.com/game/demo
I'm building on an ad-free website with 50+ solitaire/puzzle games.
Gotten some feedback from HN already and now fixing things – basically rewriting the engine for the 4th time.
Hoping to add some hundred games more soon!
1. I don't like how if I already place an Ace and then reveal a 2 that matches it suit, it then automatically moves it over. While mostly trivial, it still takes a way one move from the player and automates it.
2. Would you be willing to expose the value that seeds the game? That way any particular game of solitaire is replayable or sharable with a friend.
https://github.com/nlopes/acdc
Still lots to do but I've been having a lot of fun.
Will look up his book.
Games for Guitarists.