Posted by david927 3 days ago
Ask HN: What Are You Working On? (Nov 2025)
Automatic dataset builder and cleaner. (End product would be something like, you type that you want a dataset for dog breeds, and it builds and cleans the data automatically)
A flashcard app for my main stakeholder (My wife)
One thing I think may be cool, is put a thumbs up/thumbs down link in the email to track sentiment of the links you share. Some links are really cool, others I am not interested in at all, it may be useful to capture that info.
I would visit a site daily if you expose some of that info publicly (like 49% positive 51% negative for a link), to see how my sentiment matches your wider audience.
It's called Stagehand (https://github.com/browserbase/stagehand) and we just released v3, which is a total rewrite.
I rewrote Playwright to run completely in a Chrome Extension without CDP or chrome.devtools for no practical reason at all. I started to do it like Forest Gump started running. It can't get past bot protection so pretty worthless from a browser automation point of view. [0]
What I don't understand is why the need to rewrite Playwright instead of just patching it. Playwright (or Puppeteer) has addressed every edge case that has come -- especially race conditions which are a monster to deal with -- up over the years and by the time you do the same you will have Playwright.
Why is rewriting or rebuilding Playwright from the ground up needed?
[0] https://github.com/adam-s/cordyceps/tree/main/pages/side-pan...
Early use case is replacing API keys with identity tokens that expire, delegate, and prove possession and then can be used for easy step up to fine-grained authorization. There's some pretty interesting authorization stuff you can do, like having multiple parties sign off before a token is valid or requiring a series of micro-services sign a token for it to be valid.
I'm currently working on Dice of Sending - for when you want to roll physical dice but you play DND online. This is mostly just for fun with my DND group.
Market is brutal though man. She hasnt gotten an offer after so much trying
The real goal is to figure out how to use code gen AI (Cursor) effectively for data science projects and to figure out rapid deployment. I'm pushing things a bit harder than you typically see in demo apps (e.g. different chart types (e.g. violin plots, heatmaps, line charts), interactive charts, JavaScript widgets interacting with Bokeh charts, etc).
I'm trying to figure out all the skills, processes, and training you need to build a technical app very quickly. I'm at the deployment stage now.
I am pretty sure I can get 70% predation rates +/- 10% . Unfortunately, I'm blocked by the lack of hardware. Kind of not-quite school affiliated (so I cannot really ask for national computing resources), so I am trying to build a single threaripper pro node on my own. Hurts the wallet, but if added to slurm as module, this can have implications.
We're at genevabm.com if you want to check it out!
I'm happy where it's landing so far but also appreciate any actionable feedback to make it better (!). Under the hood, it packs a Rust Axum API, plenty of ffmpeg, and some hobo infrastructure [2] here and there.
[1] - https://nid.nogg.dev