01 / Definitions
One metric, many answers
Revenue, customer, and channel can each mean several things. If the approved definition lives in a dashboard, a wiki, or someone's memory, an agent has to guess.
Semantic layer for AI
Make your business data usable by AI.
We organize your data so AI knows what it means — and how to use it.
No pitch deck — a 30-minute technical conversation.
- Practice
- Founder-led
- Current work
- Architecting the semantic layer behind agentic analytics at a Fortune 500 consumer goods company
- Experience
- A decade across data science, analytics, and data products
Initial acquisition focus · Broader semantic foundation
Is your Power BI model ready for Copilot—or only for dashboards?
Start with one consequential question. Refinity examines the model, definitions, AI context, security, and evaluation evidence needed to make the answer defensible.
The semantic gap
Your AI can query the data. It still doesn't know what the data means.
Tables expose fields. They do not explain definitions, grain, relationships, and ownership. Refinity turns that missing business context into a semantic contract both people and agents can use.
02 / Context
Context is a contract
A longer prompt cannot repair ambiguous data. Agents need explicit grain, relationships, business rules, and ownership close to the data they query.
03 / Evaluation
Trust must be testable
A plausible answer is not enough. Every important claim needs a source, an expected result, and a quality gate that catches drift before a decision does.
What Refinity builds
The layer between your data and your AI.
Refinity designs the business context, governed access, and quality gates that turn raw enterprise data into dependable agent answers.
- 01Semantic models that encode the business definition
- 02Agent-readable context tied to governed data
- 03Evaluations that make answers checkable
Representative engagement patterns
The shapes of work we take on.
Semantic layer at scale
A BI estate rebuilt on a governed semantic layer
Migrating enterprise reporting onto dimensional models sized against real capacity limits — so the layer survives production query loads, not just the demo dataset.
Agent-ready analytics
Giving AI agents governed access to the numbers
Structured, governed agent access to semantic models — with evaluation harnesses gating what agents are allowed to answer, so wrong answers get caught before decisions do.
BI delivery pipeline
Quality gates on every model change
Git-integrated BI development with automated best-practice checks on every pull request — a short, fixed-scope engagement that upgrades how the whole team ships.
Founder-led
One senior owner from the first question to handover.
Refinity is led by Carson Leung, a data product leader and architect with a decade across data science, analytics, and enterprise data.
About Carson and the practiceContact
Let's make your data legible to AI.
Tell me what your AI needs to understand. I'll map the semantic gap, tell you whether Refinity can close it, and outline what the work would take.
I reply within one business day.