Clarity over cleverness
If a system needs explaining before it can be used, the design is unfinished.
Alex OS OnlineAlexandra Nac
Alexandra NacSenior Full-Stack Developer | Applied AI Developer | AI Workflows · RAG Systems · Agentic Applications
What I've shipped
Built, shipped, measured.
The systems, and what they changed.
Live constellation12
Select a named star to explore the work
What I've shipped
What I've shipped
The one I'd show first
The things that exist because I built them — each one with the outcome it produced and the limits it still has.The person behind the system
I design and engineer AI systems — retrieval, reasoning, and the interfaces that make them trustworthy.
You're inside one right now: the journeys, answers and recommendations on this site come from a reasoning system I designed, directed and run. I built it the way I work — with AI as a force multiplier, and every architectural decision my own.
Core Preview
Inside the Alex OS architecture
Content Registry
CMS
A question arrives — everything starts from real, published content.What I've shipped
More of what I've built
The systems, and what they changed.
01Evidence
03Projects
See every projectContext assembly 10.5s to 1.8s
Alex OS assembled its content blocks on every request. A read-through block cache cut assembly time from 10.5s to 1.8s, verified end-to-end.
1.8s (from 10.5s)↗02Evidence0Hardcoded homepage types
Dynamic CMS proof
↗03ProjectsTokens cover color and styling rules first. Typography is newer ground — the current font direction ships as a real CMS setting with three candidates, not a settled choice.
Theme DNA Design System
↗Three systems, told as stories
The work I'd start with
ai_rag
Alex OS
ProblemI couldn't find a creative way to show AI engineering and UX in one place. A static portfolio could list my work, but it couldn't demonstrate it.
What changedAlex OS became the all-in-one answer: content, colours, themes and even the assistant's tone all live in the CMS — Theme DNA styles the site, Voice DNA shapes how Alex speaks. When I want to change anything tomorrow, I edit content, not code.
ProofSource-grounded answers02
cms_web
Alex CMS
ProblemA hand-coded content model means every new kind of content needs a schema migration and new frontend code. That doesn't scale for a site that keeps changing shape.
What changedFully dynamic content model with zero hardcoded types.
Proof15 registered content types — 10 publicThinking
Notes from inside the machine.
6 Connected disciplines
Across the stack. Never without context.
How I make decisions
The principles behind the systems
Evidence over performance
I don't want AI features that only look intelligent. I want traceable sources, visible limitations and measurable outcomes.
Systems that remain editable
I build tools I can keep shaping, rather than polished demos that collapse when requirements change.
Questions
Asked and answered
- Who is Alexandra Nac?
- I'm Alexandra Nac (Alex), a full-stack AI engineer working in RAG systems, agentic workflows, CMS architecture, UI/UX thinking, and automation. This site is my personal knowledge hub — work, field notes, custom tools, and experiments from building real systems.
- What does Alexandra Nac work on?
- Four core areas: applied AI (RAG systems, agents, prompt workflows, AI evaluation, structured LLM outputs), full-stack systems (frontend, backend, APIs, CMS architecture, data workflows), UI/UX thinking (interfaces that guide users and reduce confusion), and custom workflows (MCPs, automations, and repeatable tools for building and analyzing software).
- What is Alex OS?
- Alex OS is the reasoning layer behind this website — a separate FastAPI service that sits between the CMS and the public site. It reads the content registry, reasons over it through configurable profiles, and returns structured recommendations, answers, and actions. It's enhancement-only: if the service goes offline, the core site still works normally.
- What kind of projects has Alexandra Nac shipped?
- Selected work spans AI & RAG, CMS & web platforms, data quality, UI/UX, and automation — including a RAG knowledge platform and an agentic CMS pipeline, among systems shipped across multiple domains. See the Work section for the full list.
- How can I contact Alexandra Nac?
- Through the contact form on this site, by email, or via the GitHub and LinkedIn links in the footer.
- What tools does Alexandra Nac build?
- Custom skills, MCPs, prompt systems, workflows, and AI evaluation tools I build to improve how I build, debug, review, and work with AI — documented across the Thinking and Projects sections.
Why any of this matters
The systems I build are technical. The goal is always human.
Less confusion, better decisions, and more control.