Skip to content

Work

We show real work, not logos: systems we've actually built, the constraints we built them under, and the numbers behind them. Here's a look at what we've shipped — with more added as clients clear us to tell their story.

01Featured

SchoolAI

privacy as the starting point

An AI platform for schools, running on student data — built with privacy as the starting point.

SchoolAI is a working product, not a slide. It applies the same thesis Palanax is built on: powerful AI working on sensitive data (here, student records) without that data being handed to a third party. It's our clearest proof that the private-AI approach isn't theory — it's deployed and running.

What it demonstrates

  • Private data handling
  • Retrieval over real document sets
  • A production system maintained over time
  • A domain where data sensitivity is non-negotiable
Read the write-up

02Anonymized engagements

As client engagements complete and permissions allow, short, honest case studies appear here — the problem, what we built, what it cost, and what we'd do differently. We publish the trade-offs, not just the wins.
Write-up in Writing

Private document assistant, professional-services firm

A retrieval assistant over a firm's contract archive, deployed in their own cloud tenant. Three weeks.

Coming soon

Document extraction pipeline, operations team

Automated extraction and classification over inbound paperwork, integrated into an existing system.

03The lab

Between client work, we build in the open. Hardware experiments, model benchmarks, cost breakdowns — published with real numbers.

It's how we keep our edge sharp and how you can judge our engineering before you hire us.

Open invitation

Book a free AI Reviewhello@palanax.com

No retainer required to start. No obligation after the review.