Live data from Hacker News

Ask HN: What are you working on? (August 2026)

news.ycombinator.com

981–990 of 1001 posts

Re: Ask HN: What are you working on? (August 2026)

#983
Developing a real-time data processing platform in Go, handling approximately 40 concurrent WebSocket streams with microsecond decision latency. The most interesting part was the parallel connection of three LLM providers as a consensus layer for automated decision making, which only works when all three providers agree.

Re: Ask HN: What are you working on? (August 2026)

#986
I have been working on calculating and predicting lottery numbers and am currently publishing the results on Zenodo; however, my work is in Brazilian Portuguese—comprising 19 academic technical articles—which you can check out here: https://zenodo.org/search?q=metadata.creators.person_or_org....

Re: Ask HN: What are you working on? (August 2026)

#989
post #256

A diffusion-based UI design tool I left Figma to build. My core hypothesis is that LLMs are trained to write consistent code (good for a billing function), but they also express design through code, resulting in consistent design as well. Diffusion models are just now good enough to do UI, and are trained on a much wider set of visual data, resulting in more unique outputs. LLMs are now good enough to convert images…

I agree with your hypothesis! Starting with the visual design rather than the implementation details in order to generate delightful user interfaces has produced good results for me personally. I've come back to your site a few times in order to launch personal projects that involve UI . I do want to say, though, that you haven't made any money off of me because I'm still working on my intro credits. As a hobbyist I'm not willing to spend a ton but even so, I think you are being over-generous. You may want to look into expiring credits / monthly-minimum style subscriptions. Good luck!

Re: Ask HN: What are you working on? (August 2026)

#990
I already answered this, but responding led me down a path to a better answer both for me, and our product. I said last week that we were working (at this+that) on building and maintaining a db of operating knowledge for companies. A crisper answer is that we're solving the problem of "unrecorded knowledge." Apparently 42% of things known by a company are only known by one employee and three of our key features are designed to help with this but I hadn't connected the dots. 1) task extraction 2) automatic contact updating and 3) the one I wrote about before: maintaining a company's operating knowledge or "brain"
Post reply on HN