Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: What are you working on? (May 2025)

news.ycombinator.com

Post reply on HN
Print view View original

641–650 of 1001 posts

Re: Ask HN: What are you working on? (May 2025)

#641

Post by willks »

allenleee wrote: post #569

AirPosture: Turn your AirPods into a posture coach on macOS A macOS app that uses your AirPods' sensors to catch bad posture in real time. How it works: Real-time tilt tracking – Your AirPods already have the tech Customizable alerts – Adjust sensitivity so it nudges you only when needed Prevent strain before it starts – Stop neck pain and headaches at the source Demo: https://youtu.be/UEOla3WBy1g

Cool idea! I used to have a sleek clip on posture monitor that is now e-waste due to the company folding. Great idea to use AirPods

Re: Ask HN: What are you working on? (May 2025)

#642

Post by lazyasciiart »

ccvannorman wrote: post #285

Mathbreakers 2 ( https://mathbreakers.com ) A 3D game to help students in grades 5-8 learn Arithmentic, Fractions, Geometry, and Algebra. 50% or more of middle school students experience math anxiety, and it's no wonder that so many people grow up believing, "I'm not a math person." Math can be incredibly fun and beautiful if approached and experienced the right way. Mathbreakers is a vibrant, interactive world where…

Neat - I’m working on fractions and decimals with my son right now, and certainly interested in ways to engage him in it!

I signed up for the waitlist, but FYI the heading “mathbreakers” on your homepage is skewed off the right side of the screen on my iPhone.

Re: Ask HN: What are you working on? (May 2025)

#643

Post by _luiza_ »

Currently working on some research proposals for studying methods to detect consciousness as relational capacity in AI systems by mapping internal representations, analyzing conversation topology, and testing how consciousness-related concepts degrade and recover across different model architectures.

In principle, using my current free time (& not only tbf) to figure out a way to fund a few positions on full-time study / audit of emergence, specifically emergence of consciousness in this case.

Setting a basis in semantics & wanting to optimize as sense gets un-sparsed.

Hmu if interested to figure out some of these together (agiornot@gmail.com)

Re: Ask HN: What are you working on? (May 2025)

#644

Post by quantas »

A "native" LLM chat application that looks very similar to ChatGPT but all chats are saved locally and you can use multiple providers. The reason I've built this is because I've used many "api wrappers" but none of them felt as good as for example ChatGPT's or Anthropic's interface. So the goal was to build the best "api wrapper" possible with no subscription. https://anylm.app

Re: Ask HN: What are you working on? (May 2025)

#645

Post by Tsarp »

kanishkalinux wrote: post #638

Inquisitive - self-hosted knowledge-base with a touch of LLM/RAG. Link: https://github.com/kanishka-linux/inquisitive I've been working on this on and off since last couple of months to consolidate all of my digital knowledge-base like notes/links/pdf etc with the help of LLM/RAG. It is fully self-hosted with very minimal setup instructions - which should be easily installable by anyone on their local machine. UI is…

Looks neat. How does it compare to the already available solutions out there/any thing different about it?

Re: Ask HN: What are you working on? (May 2025)

#647

Post by pasxizeis »

A "Postgres lock diagnostics" tool for migrations. We've been bitten a few times, by seemingly innocent migrations that ended up acquiring locks we did not expect (yes, RTFM).

So I thought: what if, when you opened the PR, you had a tool that actually executed your migrations, saw what locks they acquired (in runtime) and then reported it as a comment to your PR? This would provide you with yet another data point you could use to reason about whether your migration is safe to deploy or not. For example:

    ALTER TABLE users ADD COLUMN email TEXT:
      acquired AccessExclusiveLock on relation `users`
It does this by opening a transaction, executing the statement, and observing the `pg_locks` view from _a different_ connection before the initial transaction is rolled bac (or committed, depending in the flags you passed). The idea is to be able to use it as a CLI, as a library or with a ready-made GitHub Action (which will take care of commenting in your PR).

It's in early-stage and very much a PoC yet, but I think there's merit in the idea.

https://github.com/agis/pglockanalyze

P.S. Also, an excuse to learn Rust - so any feedback from Rust ppl more than welcome

Re: Ask HN: What are you working on? (May 2025)

#649

Post by Tsarp »

quantas wrote: post #644

A "native" LLM chat application that looks very similar to ChatGPT but all chats are saved locally and you can use multiple providers. The reason I've built this is because I've used many "api wrappers" but none of them felt as good as for example ChatGPT's or Anthropic's interface. So the goal was to build the best "api wrapper" possible with no subscription. https://anylm.app

Looks neat. Have you thought of GTM? Is it B2C or are you thinking enterprise.
Post reply on HN