One permission model
Access is defined against the data, not against a copy of it.
Enterprise AI
Retrieval, context and records in one request.
Enterprise AI fails on governance before it fails on models: what may be retrieved, for whom, and from which source.
An AI feature that answers under the same access rules as the systems it reads.
AI features across an organisation’s systems, with access rules that hold.
An enterprise AI feature reads across records, documents and relationships, and its answers have to respect the same boundaries as the systems it reads. PLOMID keeps the data in one governed layer, which is a precondition for a feature that can be explained. Retrieval is designed to run over that same governed layer.
An AI feature that answers under the same access rules as the systems it reads.
Environment
The question
A user question arriving from an application, scoped before anything is read.
SQL
Two estates, drawn.
The same shapes, held two ways. The difference is not the storage, it is where the agreement between them lives.
Separate systems Copies kept in step
One layer One plan · one contract
What this workload moves.
Every shape below is a surface of the layer, not a format to be converted into one. They are read and written together.
- SQL Records, keys and joins
- JSON / documents Documents and nested objects
- Vector Similarity and semantic retrieval
- Graph Relationships and traversal
How the workload reaches the data.
The models this workload names, the path a request takes through them, and the surfaces that speak to the layer.
Enterprise AI · architecture
Workload
The shape of the work itself.
- Ask
- Scope
- Retrieve
- Answer
Data models
The models this workload names.
- SQL
- JSON / documents
- Vector
- Graph
The layer
One path from a request to the data it names.
- Planning Predicates narrow the work before it runs
- Execution Rows, fields and windows answered together
- Transactions Readers and writers do not block each other
Surfaces
How the workload reaches the layer.
- SQL surface The query language the layer is documented in
- Applications Services and jobs reading and writing as they run
- Analytics & AI clients The same layer, the same access path
The work, stage by stage.
Choose a stage to read what happens there, or a shape to see every stage that handles it. Nothing on the map is a private interface.
PLOMID · Enterprise AI ask · scope · retrieve · answer Select a stage
Stage
Ask
A user question arrives from an application or client
SQL
01
Ask
A user question arrives from an application or client
- SQL
02
Scope
Access and dataset boundaries applied first
- SQL
- JSON / documents
03
Retrieve
Candidates narrowed by similarity and structure
- Vector
- Graph
04
Answer
The answer and its sources recorded
- SQL
- JSON / documents
What changes when the data is in one place.
Stated as properties of the system rather than as results we cannot measure for you.
Sources recorded
An answer can carry the records and documents it used.
No second estate
Retrieval stops creating a parallel store with its own lifecycle.
Mixed questions
Structural conditions and similarity are part of the same request.
What this is built on.
Enterprise retrieval is a governance surface. These are the parts of the layer it depends on.
- Retrieval as an access path → Similarity search arrives as another way into the same data rather than a second copy to keep in step.
- Every client, one layer → Applications, AI clients, services and background jobs reach the same data through the same interfaces.
- One data layer → Rows, documents and time-ordered events live in one system, so a question is asked once instead of once per store.
- Relationships held as data → Edges expressed directly remove the reconstruction that happens at query time when relationships live in a join.
Workloads that sit beside this one.
Where this shows up.
Derived from the industry pages that reference this solution, so the two directions of the relationship always agree.
- Banking → Accounts, transactions, counterparties and the relationships between them.
- Risk & Fraud → Signals, relationships, cases and decisions across an institution’s data.
- Enterprise → Systems of record, documents, workflow events and reporting in one organisation.
- Supply Chain → Suppliers, orders, inventory, movements and the documents between them.