Selected work

Applied AI · Knowledge retrieval · Monitoring

Compliance AI

An AI answer needs evidence people can inspect.

An AI knowledge base and regulatory monitoring for agency consultants working on client projects, with evidence available for review.

At a glance

Used by agency consultants for knowledge retrieval and monitoring legislative and regulatory changes.

In use and in active development. No formal answer-quality evaluation completed.

Contribution
Product analysis, UX & AI-assisted build
Period
May–September 2026 · Ongoing
Status
In use by agency consultants · Active development
In use
by agency consultants on client projects
End to end
stakeholder needs, UX, architecture and build
Source-aware
retrieval, access rights and citation checks

The recording shows the monitoring module. The wider app also supports questions, online research and policy review, with a choice of OpenAI and Anthropic models.

01

The challenge

Agency consultants use the application as a retrieval-based AI knowledge base and to monitor legislative and regulatory changes while working on client projects. Their research spans source material that changes over time.

The product needed to connect three questions: what changed, what does the evidence say and where does a person need to look more closely?

02

My role

I gathered stakeholder needs, wrote user stories, designed the architecture and frontend, and built the application with AI assistance. I also separated implementation tests from answer-quality evaluation.

03

The decisions

Make the source inspectable

Knowledge-base answers use stored material, respect access rights and check quotations against sources. The system can decline without evidence.

Reviewers can inspect the basis of an answer; a matching quote alone does not prove the interpretation.

Make the research mode explicit

Stored-knowledge answers and wider online research have separate modes.

People can distinguish the source boundary before assessing the response.

Define how answer quality will be checked

I created a separate framework for evaluating answer quality, alongside implementation tests. No formal answer-quality evaluation has been completed.

The framework provides a way to assess retrieval and interpretation; current use does not establish answer reliability.

FastAPIReactSupabaseHybrid retrievalQdrantOpenSearch
04

The result

The agency uses the knowledge base and monitoring application in client-project work. No formal answer-quality evaluation has been completed, and no accuracy or time-saving result is claimed.

Development continues on scanned-PDF support, retrieval ranking and source coverage. The recording shows the monitoring module.

What I’d measure next
Measure
Answer correctness, source support and appropriate refusal.
How
Build a representative, expert-reviewed question set, including unanswerable cases. Score correctness and whether cited sources support each claim. Report unsupported claims, missed refusals, source coverage and time for a reviewer to verify an answer.
Check alongside it
Evaluate access-rights failures separately. Deployment and citation matching are not proof of reliable answers.

Explore more work.

11 projects

Scroll to explore all projects.

Pavol JuríkProduct / UX / AI

Let’s talk

Good things start
with a conversation.

Have a role, a project, or an idea in mind?
Tell me a little about it.