Case Study
How Sella Delivers Scalable Credit Risk Explanations with Prometheux
Read More

The knowledge graph platform
built for on‑prem AI

Run the data processing your AI needs on the system that keeps you in charge.

Speak to us

Trusted by technical leaders in regulated enterprises

Other graph platforms tie your hands

Limited room to scale

Graph platforms can’t distribute workloads across machines, so you’ll spend more on bigger servers

Too much custom code

They only handle nodes and edges, so you’ll need other tools for the rest of the workflow

Second data stack required

Copying data into a dedicated graph database requires money, migration, and maintenance.

Choose Prometheux

Stay in control with proprietary knowledge graphs

30+ sources

queried in one unified knowledge graph

9x faster

graph reasoning

<1 min

per regulatory flow

+42%

retrieval improvement

Fortune 100 retailerRead →
€9M

saved per year

Financial institutionRead →

Reveal what your
data implies

Our living knowledge graph augments itself by deducing hidden knowledge. We don’t just look at what the data says — we derive new facts from existing information.

“Prometheux is more than just technology; it is a catalyst for fact-based decision-making. The ability to query heterogeneous databases through a flexible ontology makes it possible to infer deep knowledge and automate complex tasks, delivering trustworthy insights that far exceed the capabilities of traditional systems.”
Anna FavaSenior Manager, Data & AI · Dompé
ReasoningCompany control · central bank
controls · 60% controls · 30% + 30% = 60%
stake(X,Y,S)  <- own(X,Y,S).
stake(X,Y,S)  <- controls(X,Z), own(Z,Y,S).
total(X,Y,T)  <- stake(X,Y,S), T = msum(S).
controls(X,Y) <- total(X,Y,T), T > 0.5.

+ controls(HoldCo, Alba)60% held directly

+ controls(HoldCo, Borea)30% direct + 30% through Alba, which it controls

Run complex reasoning
in just one engine

Write in any language: Python, Cypher, Vadalog, Datalog, SQL, natural language, and more. Query tables as graphs, graphs as tables, or any combination — all in one place. Data never gets locked in again.

See how we handle ontologies >>
“Prometheux’s research-backed platform abstracts away our various data sources, letting us iterate quickly […] instead of writing complex database queries or migrating data. […] The explain functionality […] is extremely valuable in the financial sector where audit trail and logical transparency is of very high priority.”
Group Head of Advanced Risk AnalyticsLeading neobank
at_risk⧉ API▶ Run
depends_on(P,S) <- supplies(S,P).
depends_on(P,S) <- depends_on(P,M), supplies(S,M).
at_risk(P) <- depends_on(P,S), disrupted(S), not second_source(P).
WITH RECURSIVE depends_on AS (SELECT plant, supplier FROM supplies
  UNION SELECT d.plant, s.supplier FROM depends_on d JOIN supplies s ON s.plant = d.supplier)
SELECT DISTINCT plant FROM depends_on JOIN disrupted USING (supplier)
WHERE plant NOT IN (SELECT plant FROM second_source);
MATCH (s:Supplier {disrupted: true})-[:SUPPLIES*1..]->(p:Plant)
WHERE NOT (p)<-[:COVERS]-(:SecondSource)
RETURN DISTINCT p.name AS plant
hit = depends_on.merge(disrupted, on="supplier")
covered = hit.plant.isin(second_source.plant)
at_risk = hit[~covered][["plant"]].drop_duplicates()
Which plants are at risk if a supplier at any tier is disrupted,
unless an approved second source covers the part?

✓ 3 rows · same result in Logic

plantvia
LyonFastenix → Kora Steel
TurinKora Steel
GdańskFastenix → Kora Steel

Handle massive workloads wherever you choose

Distribute and parallelize graph traversal and data processing at peak efficiency and lower cost. Our proprietary engine handles sub-second queries and petabyte-scale processing.

See our benchmarks >>
“With Prometheux, we achieved a speed-up of over 250% on critical tasks involving complex reasoning over knowledge graphs with hundreds of millions of nodes and billions of edges.”
Head of R&DFinancial institution
Engine
Without Prometheux
Edges traversed0Facts derived0running…
With Prometheux
Edges traversed0Facts derived0running…
WithoutWith Prometheux
Time to finish
Compute cost

Build around your
existing infrastructure

Deploy anywhere, and never get locked into a specific database, cloud, or AI model. Your knowledge graphs stay portable and run on Snowflake or Databricks.

“Prometheux enabled us to build a system that can access our databases, process all the client’s historical variables, and provide an intelligent, understandable explanation autonomously.”
Massimiliano MalinHead of Credit Risk · Banca Sella
Architecture
AnalyticsGraph and BI
IntegrationsBack into your apps
AgentsRun workflows
Runs wherever you choose SaaS
Your data, where it is · any source
PostgreSQLOracleMongoDBSnowflakeDatabricksBigQueryTeradataMariaDBMySQLSAPNeo4jParquetSQLitePrestoHubSpotShopifyStripeZendeskJiraNotionGoogle SheetsAirtableGitHubQdrantFiles

See the platform

Connect your data, encode your rules and trace every answer, then put it to work in apps and agents. All in one place, on your infrastructure.

Book a demo

Click any tab, or let it play.

Request access

Tell us who you are and we’ll get you into the platform.

Prometheux supply_risk

AI

Any MCP agent
You

Which plants are at risk if Kora Steel halts output?

prometheux · reason at_risk

3 plants are at risk.

  • Lyon · tier 2, via Fastenix
  • Turin · direct supplier
  • Gdańsk · tier 2, via Fastenix

Same question, same answer, every time.

Start building your knowledge graph, or ask about your data, rules, results…

Data

supply_risk5 tables

6 columns · read in place

supplier_idstringnamestringcountrystringtierintrisk_scorefloatupdated_attimestamp

6 columns · read in place

part_idstringsupplier_idstringplant_idstringqtyintunit_costdecimalvalid_fromdate

5 columns · read in place

plant_idstringcitystringcountrystringcapacityintopeneddate

4 columns · read in place

part_idstringalt_supplier_idstringlead_time_daysintapprovedbool

5 columns · read in place

supplier_idstringcert_typestringissuerstringissueddateexpiresdate

Query tables as a graph, or graphs as a table, in any combination. No data movement.

at_risk
2 notes non-deterministic ⧉ API Supply chain
IndentWrap✦ Ask AI
1% A plant is at risk if a supplier at any tier is disrupted,
2% unless an approved second source covers the part.
3depends_on(P,S) <- supplies(S,P).
4depends_on(P,S) <- depends_on(P,M), supplies(S,M).
5at_risk(P) <- depends_on(P,S), disrupted(S), not second_source(P).

✓ 3 rows · same result in every language

plantvia
LyonFastenix → Kora Steel
TurinKora Steel
GdańskFastenix → Kora Steel
IndentWrap✦ Ask AI
1WITH RECURSIVE depends_on AS (
2  SELECT plant, supplier FROM supplies UNION
3  SELECT d.plant, s.supplier FROM depends_on d
4    JOIN supplies s ON s.plant = d.supplier)
5SELECT DISTINCT plant FROM depends_on JOIN disrupted USING (supplier)
6WHERE plant NOT IN (SELECT plant FROM second_source);

✓ 3 rows · same result in every language

plantvia
LyonFastenix → Kora Steel
TurinKora Steel
GdańskFastenix → Kora Steel
IndentWrap✦ Ask AI
1MATCH (s:Supplier {disrupted: true})-[:SUPPLIES*1..]->(p:Plant)
2WHERE NOT (p)<-[:COVERS]-(:SecondSource)
3RETURN DISTINCT p.name AS plant

✓ 3 rows · same result in every language

plantvia
LyonFastenix → Kora Steel
TurinKora Steel
GdańskFastenix → Kora Steel
IndentWrap✦ Ask AI
1def at_risk(depends_on, disrupted, second_source):
2    hit = depends_on.merge(disrupted, on="supplier")
3    covered = hit.plant.isin(second_source.plant)
4    return hit[~covered][["plant"]].drop_duplicates()

✓ 3 rows · same result in every language

plantvia
LyonFastenix → Kora Steel
TurinKora Steel
GdańskFastenix → Kora Steel
ModeStatic Pin a fixed set of notes

✓ 2 notes pinned · read by at_risk and by every agent

IndentWrap✦ Ask AI
1% Explain each at-risk plant for the operations team.
2% Input: at_risk(plant, via). Output: plant, explanation.
3Explain in one sentence why {plant} is at risk,
4naming the supplier chain {via}. Plain language, no jargon.

✓ 3 rows · one explanation per at_risk plant

plantexplanation
LyonBuys from Fastenix, which depends on Kora Steel.
TurinBuys directly from Kora Steel, with no approved second source.
GdańskBuys from Fastenix, which depends on Kora Steel.

Natural language

A plant is at risk if a supplier at any tier is disrupted, unless an approved second source covers the part. Tiers are counted through the bill of materials.

NoteAt risk means any tierA disrupted supplier reaches the plant through any number of tiers.
at_risk
NoteOnly approved second sources countPending approvals do not cover a part.

The notes this concept reads, and every agent sees.

Engine activityEngine online● Engine active · 0 running · 1 worker
OverviewJobs 42Executors 1RuntimeLogsStatus

Session

Started at09:00 today
Total uptime2h 14m
Runs onyour premises

Tasks 92% complete

Done169
Running0
Failed0
Total184

3 results · executed in 1.6 s

at_risk
plantviatier
LyonFastenix → Kora Steel2
TurinKora Steel1
GdańskFastenix → Kora Steel2
5 conceptsEngine
GroupsSupply network 3Risk 2
suppliers
bill_of_materials
second_sources
disrupted
Python▥ 1 rows◷ 0.8 s
supplies
SQL▥ 128 rows◷ 1.9 s
second_source
Logic▥ 12 rows◷ 1.1 s
depends_on
Logic▥ 342 rows◷ 2.4 s
at_risk
Logic▥ 3 rows◷ 1.6 s
at_riskLogic⧉ API

Natural language

Fields


                    

✓ at_risk traced back through 4 concepts to 3 sources

← ModelConcept graphsupplies

10 nodes · 9 relationships · read in place

SUPPLIESSUPPLIESSUPPLIESSUPPLIES Orex MiningKora SteelFastenixBrackwellAlloyworksLyonTurinGdańskVelo PolymersBrno
Building graph…
Tier-2 supplierSupplierPlant
nodetypescore

Apps Supply risk monitor

3plants at risk
1supplier disrupted
0copies of your data
plantexposurestatus
Lyontier 2 · FastenixAt risk
TurindirectAt risk
Gdańsktier 2 · FastenixAt risk

Exposure graph

Kora SteelFastenixGdańskLyonTurin

✓ Live for your team · web app, API and MCP

Business users get the same answers as your agents.

Prometheux lets you rely on your agents

In regulated industries, AI must be reliable, governed and has to integrate seamlessly within existing systems to get to work, fast.

risk-desk — claude
✻ Welcome to Claude Code!
cwd: ~/risk-desk · MCP: prometheux connected
> which of our counterparties are controlled by a sanctioned entity?
 
⏺ prometheux - reason (MCP)(ontology: "ownership", query: "flag(C)")
⎿ 4 rows · deterministic · 1.4s
⏺ 4 counterparties have a sanctioned ultimate controller:
• Nordhavn Capital 3-hop ownership chain
• Brisa Logistica SpA 2-hop ownership chain
• Kestrel Trade Ltd joint control, 2 holders
• Orvelle Metals direct 54% stake
 
> why is Nordhavn Capital flagged?
 
⏺ prometheux - explain (MCP)(fact: "flag(Nordhavn Capital)")
⎿ owns(Varko, Eastline BV, 0.62) ← snowflake #88213
owns(Eastline BV, Nordhavn, 0.70) ← postgres #1204
sanctioned(Varko) ← s3 sdn_list.csv:5531
>
? for shortcuts⏵⏵ auto-accept edits on
Any MCP-capable agent
Claude, ChatGPT, Cursor, Copilot, or straight from the terminal
Docs ↗

MCP server

Agents discover your concepts and rules, run reasoning and fetch explanations through Model Context Protocol. Access is scoped to what each user can see.

MCP docs →

API & SDK

Call the same knowledge graph from services, notebooks and pipelines over REST or Python. One definition serves apps and agents alike.

API docs →
>_

CLI

Stream results, inspect query plans and open derivations without leaving the terminal. Coding agents call it directly, with no server to set up.

CLI docs →

Get both portability and choice

Traditional approach
3Extra copies
3ETL pipelines
+2Stores to lock into
Your data, in silos
  • Snowflake
  • Databricks
  • BigQuery
  • PostgreSQL
  • Oracle
  • Teradata
Relational

ETL pipeline

Extract, transform, load

AirflowdbtInformaticaTalend+ more
SchedulerMonitoringSchema fixes
Copy 1

Marts and reporting tables

Access rulesBackupsStorage

SQL + Python

Business logic in queries and scripts

Sync
Graph

Graph loading

Export, model, load

Graph modelReloadsSchema fixes
Copy 2

Graph database

Another vendor store to run

Neo4jArangoDBTigerGraphNeptune
Access rulesBackupsSizing

Cypher

Graph logic in another language

Copy 3 · Semantic stack

Ontology and triple store

Map both into RDF, model it in OWL, query it in SPARQL

RDFoxGraphDBStardogProtégé
OWLRDFSPARQLSHACLR2RML mappings
Semantic layer

Metrics and terms defined again for BI

dbtCubeLookerAtScale

Custom dashboards and glue code

Hand-built to merge answers that can disagree

Custom dashboardsGlue scriptsOne-off agent toolsReport exports
Runs where each tool decidesThree estates to run and pay for
Relational stackYour warehouse, plus an ETL tool and its scheduler
Graph stackThe graph vendor's cloud, or a cluster sized just for it
Semantic stackA triple store and a BI semantic layer, each hosted on its own

Locked in. Your graph and semantic data live in vendor stores. Leaving means another migration.

With Prometheux
0Extra copies
0ETL
0Lock-in
Your data, where it is
Read in place
and write back
Prometheux

Graph, SQL, business context and rules, in one fully portable layer.

AI assistant

Builds your model and rules from natural language, and suggests relationships to speed things up. You review.

Runs wherever you choose Prometheux Cloud AWSAzureGCP SnowflakeDatabricksPalantir On-prem
Power deterministic
systems

Analytics

Graph and BI

Integrations

Back into your apps

Agents

Run workflows deterministically

Get value in <5 minutes

  1. 1Connect
    Point Prometheux at a database, warehouse or file. Your data stays where it is.
  2. 2Define
    Pick the concepts and relationships that matter, or let the assistant draft them.
  3. 3Encode
    Write your business logic as rules, in Natural Language, SQL, Cypher, Python or logic.
  4. 4Run
    New facts are derived from your rules, and every result can be explained.
  5. 5Use
    Query it from your apps and dashboards, or from any MCP agent.

Enterprise & complex deployments

Work closely with our Forward Deployed Engineers, from months to days.

Bring us one critical workflow. We map your data, rules and exceptions with your team, then run it on your infrastructure.

Chat to our Engineers

Working with

Own your intelligence with Prometheux

Start with one workflow today.
Scale your entire operations tomorrow.

FAQs

Is Prometheux a graph database?

No. It’s a knowledge graph platform with its own reasoning engine. You query your existing data as a graph without building one or copying it into a separate store.

Do I have to move my data?

No. Prometheux connects to SQL databases, NoSQL stores, data lakes, REST APIs and more, and processes data where it lives. No migration, no duplicate copy to keep in sync.

Where does it run?

Wherever you need it: on your premises, in your own AWS, Azure or GCP account, as a native app on Snowflake or Databricks, or as SaaS, fully managed by Prometheux.

Where do the business rules come from?

You or an agent define concepts, relationships and logic once, in plain language or code, as reusable rules that run across all your connected data. You review every rule, and our engineers can help with the first ones.

How is it different from a graph database or a semantic layer?

A graph database needs its own copy of your data, kept in sync. A semantic layer gives your metrics consistent definitions. Prometheux queries the data you already have as a graph, and your business rules derive what no table stores, like which supplier a plant depends on three tiers down. Every answer is deterministic and comes with its explanation.

Can it meet our security and audit requirements?

Yes. Prometheux is SOC 2 Type II audited, and we’ll share the report on request. Deploy it where your data has to stay, including fully on-prem. Each user’s storage is isolated, agents act with the user’s own access rights, and every result can be followed back through the concepts and sources it depends on.

How do AI agents use it?

Through MCP, the API or the CLI. Agents query your knowledge graph and get answers they can explain, instead of guessing from raw tables.

We use some essential cookies to make this site work. By clicking “Accept All” you agree to the storing of cookies on your device to enhance site navigation, analyse site usage and assist our marketing efforts. For more detailed information see our Cookie Notice.

Accept AllDecline