Oriental Nippon Development
Back to portfolio
Case studyUS revenue-cycle operator

Ask Your Data in Plain Language, a 30B LLM Running Entirely On-Premise

Business staff couldn't write SQL, so every question became a ticket. We deployed a 30-billion-parameter model on the client's own server, grounded on its schema, answering voice or text questions in seconds, with nothing leaving the network.

Industry
Revenue Cycle & Finance
Region
United States
Engagement
Lab-type ODC (dedicated team)

Outcomes

  • Answers in seconds, no SQL
  • 100% on-premise, zero data egress
  • Permission-scoped, read-only

Stack

On-prem LLM (30B) · SQL · Schema grounding · Voice input · Private server

Written byOriental Nippon Team

A US revenue-cycle operator holds account-level data on healthcare patients. Operations and commercial staff needed answers from that data all the time, but they could not write SQL, and the data team's queue was the only way in.

Hosted AI tools were not an option. The data includes personally identifiable and health information, and it was not allowed to leave the building.

The problem

  • Every question was a ticket. Ad-hoc questions waited in the data team's queue.
  • Capacity drained. The data team spent its time on one-off queries.
  • Dashboards miss the unexpected. Self-service BI covers the questions you anticipate. The important ones are often the ones you didn't.
  • Strict data residency. Nothing could cross the network boundary.
BeforeAfter
How staff askedTicket to the data teamPlain language, voice or text
Time to answerHours to daysSeconds
Model hostingn/a30B-parameter model, on-premise
Data leaving the networkn/aNone

What we built

  • A locally hosted LLM

    A thirty-billion-parameter model on the client's own server, with nothing crossing the network boundary.

  • Schema grounding

    Questions are translated into queries against the real database structure, not a guessed one.

  • A voice and text interface

    Results in seconds.

  • Read-only, permission-scoped access

    Users only see what they are already allowed to see.

A single question, end to end:

Technology and the role it played

TechnologyRole in this project
30B-parameter open modelTurns plain-language questions into queries, on the client's server
Schema groundingGives the model the real tables and columns
SQL (read-only)Answers questions without the ability to change data
Voice inputLets staff ask without typing
Private serverKeeps every prompt and answer inside the network

How we delivered it

Our ODC team settled the boundary first: what may run where, confirmed with the client's compliance position before any model was chosen.

  1. Boundary and model choice. Residency constraints, model selection and hardware sizing.
  2. Infrastructure. On-premise deployment, serving and monitoring.
  3. Schema grounding. Iterated against real questions staff had previously sent to the data team.
  4. Constrain the interface. Read paths only, scoped to the asking user's permissions.
  5. Pilot. Non-technical users, with answers compared to what the data team would have given.
Delivery timelineIndicative total: 10–14 weeks
  1. 01Boundary and model selection

    1–2 wks

    Residency constraints, model choice, hardware sizing

  2. 02Infrastructure

    2 wks

    On-premise deployment, serving, monitoring

  3. 03Schema grounding

    3–4 wks

    Structure mapping, query generation, evaluation set

  4. 04Interface and permissions

    2–3 wks

    Voice and text input, read-only scoping, user permissions

  5. 05Pilot

    2–3 wks

    Non-technical users, answers compared against the data team's

Risks and how we handled them

RiskHow we handled it
Sensitive data leaving the networkResidency boundary agreed with compliance before choosing a model
The model invents columnsGrounding iterated against real past questions
Users seeing data they shouldn'tRead-only queries run with the asking user's permissions
Answers nobody can checkPilot answers compared against what the data team would have produced

Outcome

Business users get account-level answers in plain language, in seconds, without SQL and without a queue. No sensitive data leaves the client's network to make that possible.

Data residency was the reason a hosted approach was never viable. Building for it from day one is far cheaper than retrofitting it after a pilot.

What the client keeps

  • The deployment and its configuration
  • Schema grounding artefacts
  • The evaluation question set
  • Permission scoping rules
  • Infrastructure documentation

Sources and further reading

  1. Top 10 for Large Language Model Applications, OWASP
  2. AI Risk Management Framework, NIST
  3. BIRD: a big benchmark for text-to-SQL, BIRD-bench
  4. vLLM documentation, vLLM project

Facing something similar? This is the kind of work our AI, Data Science & Automation team delivers every day. Talk to us about your stack.