Intelligence

Tequery

Querying company data without knowing the database.

Plain-language queries · read-only access · on-premise install · works with different language models

Ask for a demo How it works

The user asks

“How many accounts opened last quarter still have incomplete anti-money-laundering documentation?”

Query generated and run

SELECT a.branch, COUNT(*)
FROM accounts a JOIN kyc k ON k.account_id = a.id
WHERE a.opened_on >= … AND k.status <> 'COMPLETE'
GROUP BY a.branch

Illustrative example. The query depends on the database structure and on what the application being queried holds.

Read-only accessqueries never change the data

Traceabilityquestion, query and user are recorded

On-premise installthe data stays inside the boundary the company sets

Works with different modelscloud models or models installed in-house

The context

Getting at the data usually takes a specialist.

The information the business needs is usually already in the systems, but finding it means knowing the applications, the databases and how they relate.

Every request goes through IT

Most requests for information need a technical team to find the right data and run the query.

The knowledge sits with a handful of people

How the applications and databases are built is usually known only to a few specialists.

Becoming self-sufficient takes time

Before you can search on your own you have to learn the naming, the codes and the relationships between the data.

Hard to retrace how a figure was produced

When a result is shared without the query behind it, checking where it came from — or reproducing it later — becomes difficult.

How it works

From the question to an answer you can check.

Tequery reads a request written in plain language, turns it into a query the database understands, and returns the result together with everything needed to check what was actually run.

The record

Every answer stays checkable.

Each request can be tied to the query that ran, the user who asked and the moment it happened, which is what verification and control need.

Simpler access to the information

Information can be searched without knowing the technical structure of the database.

Requests in plain language
Less dependence on technical support for the simpler queries
New users productive sooner

Operations view

Results documented and checkable

Every answer can be verified through the query generated and the information recorded while it ran.

A record of the queries run
Results reproducible later
Support for audit and compliance

Compliance view

The infrastructure stays under your control

The language model, how it is installed and how it runs are all decided according to the organisation’s own technology policy.

On-premise install
Works with different language models
Uses the access profiles already defined

IT view

The boundary

What the system does, and what you do.

Aspect	What Tequery does	What stays with you
Data	Reads, never writes	The applications and their data stay where they are
Permissions	Respects the access profiles already in place	Managing the user accounts
Data model	Maps the questions onto the real structures	Ownership of the model and the applications
LLM	Works with different models, cloud or in-house	Choosing the model
Tracking	Records the question, the query and the user	The retention and audit policies

Frequently asked questions

The questions we are asked
most about Tequery.

Any other questions?

Tequery installs on-premise and reads the data without writing to it. Where the requests are processed depends on the language model: with a model installed in-house nothing leaves the company boundary; with a cloud model the boundary and the processing rules are agreed with IT and compliance.

Every answer is shown together with the query that produced it, so the result can be checked instead of taken on trust. The queries are read-only, so a mistake produces at worst a figure to correct, never a changed record.

No: access is read-only by design, and it respects the authorisation profiles already in place. Through Tequery a user sees exactly what they would be allowed to see anywhere else.

No. Tequery is a layer on top of the applications you already have: nothing has to be moved. If you then want a unified data platform, that is a project of its own — the ground covered by frAIme — but it is not a prerequisite.

Tequery respects the authorisation profiles already defined, and governance stays with you: which data can be queried, which profiles are enabled and how long the record is kept are all configured with IT and compliance before go-live.

Keep exploring

The other Add Value products

Intelligence · Raw materials

PriceSense

Reads news, events and market data to produce estimates of raw-material prices and availability, validated by an analyst.

Data · Cloud and AI

frAIme

Integrates data from different systems through a data fabric, making it available to analytics and AI without necessarily moving it from where it lives.

Intelligence · Mainframe

OptiMIPS

Analyses mainframe consumption, attributes it to individual applications and identifies where resources can be optimised.

Intelligence · Procurement

TSV

Categorises indirect spend with AI and makes it searchable by category and supplier.

Intelligence · Insurance

Calamìta

Cross-references seismic, weather and hydrogeological data with geolocated policies to analyse the NatCat exposure of an insurance portfolio.

Software · Modernisation

L2M

Core applications, modernised and verified against real data.

Intelligence · AI adoption

Officina AI

Finding where artificial intelligence can create value in business processes.

Reach the data without filing a technical request.

Ask for a demo All the products