Auditable Synthetic Data,
Built on Contracts
A core engine for synthetic data with auditable outputs. Define your data contract, choose a generation mode, and receive projections alongside a comprehensive evidence bundle — proving quality, not just claiming it.
How SynthLabTech Works
Define a contract. Run the engine. Receive synthetic data and an evidence bundle that proves its quality.
Define Your Data Contract
Specify your constraints, schema, and requirements upfront. The contract governs exactly what the engine produces — no surprises, no unchecked outputs.
- Schema and relational constraints defined declaratively
- Feasibility-conditional hard guarantees: zero violations for feasible contracts
- Choose determinism mode: STRICT for exact replay, FAST for speed
Run the Engine
The engine enforces your contract through deterministic relational closure. It generates projections that satisfy all specified constraints — with feasibility certificates when contracts are achievable.
- Zero hard relational violations for feasible contracts
- Reproducible runs with auditable hashes
- Scenario and regime-aware synthesis for complex systems
Receive Projections + Evidence
Every run produces two outputs: the synthetic data (projections) and a comprehensive evidence bundle. The bundle contains constraint, determinism, utility, and privacy risk reports — so you can audit exactly what was generated and how.
- Constraint Report: which rules were enforced, which were relaxed
- Determinism Report: reproducibility hashes and seed state
- Utility Report: statistical fidelity and distribution quality
- Privacy Risk Report: measured and reported, not assumed
Four Modes, One Engine
Choose the right mode for your use case — from fast tabular synthesis to full ICS security validation datasets.
Rapid Mode (RRF)
Fast Baseline SynthesisFast baseline synthesis for common tabular use cases. Schema inference, type profiling, quality reports, and baseline synthesizers — focused on speed and common workflows.
Research Mode (TRC)
World-State SynthesisWorld-state-first synthesis with relational closure, regime modelling, and comprehensive evidence bundles. For strict data integrity requirements and complex relational systems.
Virtual SCADA Mode
Scenario-Based StreamingScenario-based generation and streaming of synthetic telemetry. Sensors are contract-defined via scenario templates — deterministic replay with protocol-conformant output streams.
ICS Security Mode
SOC Validation DatasetsSecurity training and SOC validation datasets based on ICS process truth. Outputs include logs, alerts, laddered attack chains, and both benign and adversarial traces with evaluation reports.
Evidence Bundles
Every generation run produces a structured evidence bundle — a product primitive that proves the quality and integrity of your synthetic data.
Constraint Report
Documents which constraints were enforced, which were relaxed, and the feasibility status of the contract.
Determinism Report
Reproducibility hashes, seed state, and run metadata so you can replay any generation exactly.
Utility Report
Statistical fidelity metrics — distribution quality, correlation preservation, and edge case coverage.
Privacy Risk Report
Privacy measured and reported. No absolute claims — quantified risk assessment with methodology.
Core Capabilities
Relational Closure
Zero hard relational violations for feasible contracts, enforced by deterministic relational closure.
Evidence Bundles
Every run produces auditable proof: constraint, determinism, utility, and privacy risk reports.
Determinism Modes
STRICT mode for exact reproducibility with auditable hashes. FAST mode when speed matters more than bit-exact replay.
Scenario & Regime Engine
Scenario and regime-aware synthesis for industrial and complex relational systems, with explicit transitions.
Reproducible Runs
Reproducible runs with auditable hashes and evidence bundles. Replay any generation for QA, validation, or security.
Privacy by Evidence
Privacy measured and reported — evidence over claims. No absolute guarantees; quantified risk assessment instead.
What We Do Not Claim
Transparency is part of our product. We believe in honest positioning.
Fidelity vs Integrity vs Truth
Most synthetic data tools optimise for statistical fidelity alone. SynthLabTech distinguishes between three qualities — and lets you decide which matters most for your use case.
Fidelity
Statistical resemblance to real-world distributions
Integrity
Relational and constraint correctness across tables
Truth
Auditable evidence that proves what was generated and how