Synthetic data you can prove.
Realistic data for AI, software testing, healthcare and industrial systems, generated or simulated on request when the real data is sensitive, restricted or does not exist. Describe what you need, or bring a dataset; every run is validated row by row and sealed into evidence you can check offline.
RadMah AI is built by RadMah, Inc., an ITLOX company.
Free account. No card. Business or university email.
NVIDIA Inception Program member
Four teams, one problem.
They need realistic data, and they need to be able to say where it came from.
Data and ML teams
Training and evaluation data when the production tables are restricted or too small.
Synthetic dataSoftware and data engineering
Deterministic fixtures that make a test suite diffable instead of flaky.
Mock dataOT, industrial and security teams
Plant telemetry and labelled attack scenarios without touching a live process.
Industrial simulationRegulated and healthcare teams
Realistic clinical structures, and a record of exactly how each dataset was produced.
Healthcare FHIRThree things it does. One contract, one evidence chain.
A simulated plant run can feed a synthesis job; the agent can drive either. They are capabilities of one system, not separate tools you integrate yourself.
Generate
Create realistic data from a prompt, a schema, or a dataset you already have.
Simulate
Produce operational data and scenarios that no ordinary dataset contains.
Verify and orchestrate
Check what was produced, prove how it was produced, and drive all of it from your own code.
The same run, from the command line.
Everything the browser does, the rady CLI and the Python SDK do too: a reviewed contract you approve before anything is generated, a sealed evidence bundle, and a verification that runs offline with no API key. This is a real run — the contract, the rows, the identifiers and the hashes below all came from it.
| column | type | kind | values |
|---|---|---|---|
| work_order_id | int | sequence | |
| asset_tag | str | sampled | composed · 18 parts |
| transformer_id | str | sampled | composed · 7 parts |
| fault_code | str | sampled | 7 values · INSULATION_FAILURE, OVERHEATING, … |
| priority | str | sampled | 4 values · Critical, High, … |
| opened_timestamp | datetime | sampled | |
| closed_timestamp | datetime | sampled | |
| technician | str | sampled | composed · 8 parts |
| parts_cost_eur | float | sampled |
| work_order_id | asset_tag | transformer_id | fault_code | priority | opened_timestamp | closed_timestamp | technician | parts_cost_eur |
|---|---|---|---|---|---|---|---|---|
| 100001 | AST-S02-10 | TX-B-04 | OIL_LEAK | Low | 2025-07-28T19:32:00 | 2025-07-28T20:30:00 | Jamie Reed | 212.79 |
| 100002 | AST-S01-05 | TX-C-03 | INSULATION_FAILURE | High | 2025-07-18T03:59:00 | 2025-07-20T00:13:00 | Drew Patel | 116.75 |
| 100003 | AST-N04-02 | TX-B-03 | ALARM_INVESTIGATION | Medium | 2025-12-06T22:05:00 | 2025-12-10T20:13:00 | Alex Patel | 102.56 |
| 100004 | AST-N03-09 | TX-A-04 | COOLING_SYSTEM | High | 2025-12-08T18:47:00 | 2025-12-12T11:45:00 | Drew Morgan | 65.89 |
| name | rows | primary |
|---|---|---|
| maintenance_work_orders | 200 | ✓ |
Real data is often the data you cannot use.
Teams need realistic data to train models, validate software and test operational systems. The data that would be most useful is frequently the data they are least able to touch: it carries personal information, it is contractually restricted, it is expensive to capture, or it describes an event nobody wants to reproduce in the real world — a plant failure, an intrusion, a rare clinical presentation.
RadMah exists to generate the data those teams need, and to make the result verifiable — so the dataset is not just plausible, but accompanied by a record of how it was produced and what was checked.
No identifier-shape values detected; dataset clean.
- Inspected rows
- 100
- Inspected attributes
- 15
- Repairs applied
- 0
- Scanner
- v1-row-level
From a spreadsheet to a substation — with receipts.
- Describeprompt, schema or source
- Generateengine or simulator
- Validateevery row, every rule
- Evidencesealed, and yours to re-check
- sealed contract
- constraint report
- artifact hashes
- determinism report
blake3 9f2c 41ab 7d05 c318 … e6b1
Generation is half the job. The other half is the record.
A quality score asks you to trust the vendor that produced it. RadMah seals each run into an evidence bundle instead: the contract that defined the job, the constraint report, an index of every artifact with its own hash, and the determinism report. A verifier ships with the SDK and checks the bundle offline — no API key, no call back to us.
- Re-check the bundle without trusting our runtime
- The same contract, seed and engine version reproduce byte-identical data
- Tamper-evident: edit any artifact and the hash comparison fails
Every run is measured against your own data.
A vendor benchmark tells you how an engine behaved on somebody else’s data. So RadMah measures yours: each synthesis run scores the delivered rows against your source — distributional similarity, downstream utility, privacy indicators and exact-copy rate — and seals the result into that job’s evidence bundle, next to the contract and seed that produced it. Fidelity depends far more on the size and shape of a table than on any headline figure, which is why we publish the measurement rather than a number.
How evidence bundles workData that only exists when a plant is running.
Most synthetic data stops at the table. RadMah extends the same contract and the same evidence chain into operational environments: a distribution substation simulated from process physics, emitting telemetry on the OT protocols your equipment actually speaks, with packet captures included — and labelled attack scenarios you could never safely run against production.
- Five protocol surfaces on the wire — Modbus TCP, OPC-UA, BACnet/IP, MQTT and DNP3 — driven by process models rather than canned CSV files
- 67 plant templates, with a CI gate that fails the build if the registry drifts
- ATT&CK for ICS scenarios with per-event ground truth
A request as ordinary as
“a distribution substation, fifteen minutes, DNP3 and Modbus, eight signals, with a transformer over-temperature alarm”
produces a simulated run with the voltages, temperatures and alarm behaviour a process model implies — not a random walk dressed up as telemetry.
Run it in our cloud, or in yours.
RadMah, Inc. is a Delaware corporation and an ITLOX company. RadMah, Inc. is a subsidiary of ITLOX, Inc.
- Managed cloud
- We run it. You pick the residency region your data stays in.
- Self-hosted
- An offline container bundle, a licence verified on your own host, and no call back to RadMah at runtime.
- Tenant isolation
- Per-tenant keys, storage prefixes and row-level filtering, exercised by gates in CI.
- Evidence you keep
- Bundles verify offline, so an audit never depends on our availability.
Bring a dataset, or just describe one.
Run a job end to end and keep what comes out of it — the data, the quality report, and the sealed evidence bundle.