HN Explorer

Show HN: Krixik – Easily sequence small/specialized AI models (pip-installable) (github.com)

23 points by lucasllinasm ↗ HN
Our initial intention was to experiment with a smartbook idea, but we couldn't find a good prototyping tool for small/specialized AI models. That rabbit hole took us through vector databases, through infra for small AI, and finally here.

We’re particularly proud of the accessibility/simplicity of our code syntax.

Krixik’s model library is limited to eleven model types (modules) and a few dozen models, but we will significantly expand it. Other enhancements, like a local client and several 3rd-party integrations, are also planned.

You require API credentials to try Krixik. You can quickly get them through this form: https://forms.gle/WZ6PegC7pcmuGNNC9

And here’s a demo video (our YouTube channel also has a couple dozen example videos in it): https://youtu.be/WpSSYLfvfdM

6 comments

[ 3.2 ms ] story [ 33.3 ms ] thread
correa_brian
This is cool. Can I bring my own vector database? I'm currently prototyping and want to experiment with different solutions. Would be ideal if I could avoid needing to implement a few different SDKs.
lucasllinasm
Not quite yet—that's something we have on the roadmap, but at present vector databases can only really be searched through by combining a text-embedder module and a vector-db module (detail for both below). Any file processed through such a pipeline will be semantically searchable.

Text embedder: https://krixik-docs.readthedocs.io/latest/modules/ai_modules...

Vector DB: https://krixik-docs.readthedocs.io/latest/modules/database_m...

Please continue submitting suggestions and requests; we'll take them very much into account as evolving our roadmap.

itsbrex
Very cool! Syntax looks very streamlined with no fluff. Looking forward to checking this out. Thanks for sharing.
lucasllinasm
Thank you! Glad to hear you're a fan of the syntax; the idea is that it be equally accessible to an experienced dev looking to save time and to a novice tackling their first-ever AI task.

If you have any further feedback later, I'll be more than grateful for it!