Posted by david927 3/30/2025
Ask HN: What are you working on? (March 2025)
To that end, I just did a show HN on a couple of my projects
What if google did not focus on hyper commercialisation? 10 Blue links, but sorted by how they answer your query- https://www.unlob.com
Can we answer questions with lesser hallucinations? A snippet cited answer engine which only picks links that focus on answering your query - https://www.unzoi.com
Product live at https://malak.vc OSS source code at https://github.com/ayinke-llc/malak
I have a use case to have a need to provision at will hyper-v instances for others to have control over fully. I've looked into proper things for things like SCVMM and Azure local and really they do suck..
So off to build a set of automation scripts to provision hyper-v inside of hyper-v attached to S2D shares and give students admin access to them and work in teams to build things on their own.
Its going to suck... I've done this exact use case using openstack, incus, proxmox and they all kinda suck...
I need project isolation, I need freedom, I need compliance.
The compliance is where I'm stuck with hyper-v due to the powers to be....
Wish me luck!
I've run the project through cline and roo. Also tried Claude 3.7 and the 1M context Gemni 2.5 pro model. I'd say Gemini is less creative. But it's still good.
I can see how it is a productivity booster. Or at least it gives you the illusion of one. Really I think best part of it is just building out detailed documentation. That's really the killer app for me.
I built it because I was frustrated with how hard it was to manage a large number of channel subscriptions using YouTube’s default interface.
My algo often gets bad, and the best recommendations for YouTube videos come from friends' suggestions. And sometimes I just want to be recommended something totally different than what my algo would know about.
Would it be possible to add that feature?
Sorry, I don’t plan to add that feature.
I'd always found web app backends to eventually need something that chained background jobs together. And inevitably, it was something bespoke without a lot of observability. It was always frustrating to maintain. And while building AI RAG pipelines, I've run into the same problems.
A few improvements to go, but from what I hear from other parents (without raising it) is that they'd take advantage of something like this.
Just 2 or 3 bugs remaining before people can start playing with both River and Freenet which hopefully means we're days away (touch wood).
[1] https://github.com/freenet/river [2] https://freenet.org/