Ask HN: We Need Help with Our Anomaly Detection Algorithm
Hey, I'm a co founder of Bomfather https://bomfather.dev/. We are in the process of adding anomaly detection to our product and this write up is a pretty deep dive into how we are thinking of approaching it.
We would really appreciate any thoughts or improvements that you guys have!
The write up is on our blog page since it has a couple of images as well as latex, and HN doesn't allow images or latex formatting.
https://bomfather.dev/blog/2025-11-24-help-with-our-algorithm/ Is this product open source/free or for-profit? I ask because if this is for-profit, help to devise a proprietary algorithm that will become a core part of it feels like something that should be done as a consulting engagement, not a HN post. And in that case, we are only about a week off of the monthly hiring posts, so I'd recommend reaching out to folks who post on the 'who wants to be hired' or the freelancer threads. Different story if this is free and open source, of course. This product is intended to be open source. We are currently building it out, but we are planning to open source it in spring 2026. Thanks for asking! Why wait? Why ask people to trust that the free work they do for you will eventually be open source in the future? Truthfully, there isn't any way to prove that we are going to OSS, but we don't want to throw out a half assed product. I agree, this isn't an ideal scenario, but we want to release our product when we have a working version. We love open source, and have been working on open source for years now. https://github.com/nathannaveen. Additionally, we have already open sourced a security tool that we were building before hand https://github.com/bomfather/minefield. Well, it seems you can either throw out a half-assed product, or wait a while and throw away your half-assed product? The odds that you're going to get free work based on a promise of future open source seems low.