Alex NovaraServices
Alex helps fictional firms review their platforms and add AI where it measurably helps.
AI assistant prototype
A working assistant over your own documents in four weeks, with an evaluation set that shows how well it answers.
- For
Small and medium firms with handbooks, archives or support answers nobody can find.
- You get
A running prototype, its evaluation results and a plan for going to production.
- Independent R&D
- Citation checker
BuiltImplemented and working
- Citation checker
Architecture review
A two-day review of a web platform: what holds, what will break first and what to change.
- For
Teams about to grow a platform or replace part of it.
- You get
A written report with prioritised recommendations and a follow-up call.
Evaluation setup
An evaluation suite for an existing AI feature, so changes to prompts and models are measured, not guessed.
- For
Teams that already ship an AI feature.
- You get
An evaluation suite in your repository, wired into continuous integration.
- Independent R&D
- Chunking benchmark
TestedTried hands-on
- Evaluation dashboard
BuiltImplemented and working
- Chunking benchmark
How I work
How a collaboration usually runs. The steps adapt to the project.
Talk it through
A short call about the problem, the systems involved and what a good outcome looks like.
Agree on the scope
A written scope with the approach, the assumptions and an estimate.
Build in small steps
Work delivered in increments, checked early and adjusted when needed.
Hand it over
Documentation, deployment or hand-over, and support as agreed.