Synthetic Data Studio &
Headless API.
Simulacra's Generative Causal AI is available through the Synthetic Data Studio or the Headless API. The Studio is an interactive platform for real-time cohort boosting, causal interventions, and scenario modeling — built for insights teams who need stakeholder-ready results on their ongoing research. The Headless API is for agencies, data teams, and agentic workflows that need a validated synthetic-data and scenario-modeling engine in their stack.
Visual canvas. Five minutes to first scenario.
Upload a CSV, Excel, or SPSS export and run real-time scenarios and interventions on your research. Answer pressing questions for out of sample populations without new fieldwork. Most consumer research and marketing users live here.
Open the Studio docs →
Headless APIOpenAPI built for agents and data pipelines.
Auth0 M2M, idempotency keys, async jobs, cleaned-schema inspection, stable error codes, and agent-discovery docs at /llms.txt.
See the API →
Generative Causal AITabular-native. Diffusion-inspired. Proprietary causal inference.
The AI engine: generative causal AI for response-structure learning, interventional inference, and refusal semantics.
Read the methodology →
Get the same results from the Synthetic Data Studio or headless API.
- 01
Upload a seed dataset
Tabular file from a study you already fielded. Automated data cleaning and type-classification on upload.
- 02
Simulacra learns the population
Tabular-native generator fits the response structure of your respondent population. Usually in ~60 seconds or less.
- 03
Generate or intervene
Boost a thin segment, condition on a target outcome mix, or run do(X = x₁).
- 04 Studio
Analyze
Real-time scenario results, automated graphs and tables, and cross-tabs — all in-platform.
- 04 API
Analyze
Scenario diagnostics, novelty audit, and methodology appendix returned programmatically.
- 05
Export
All generated data, tables, and plots available for download.
Simulate your next launch, pricing move, or concept — validated on your data.
Synthetic rows
Generate coherent records: marginals match, dependencies and causal structure are preserved, and the generated population maintains empirical variance.
Conditional generations
Desired-outcome targets across categorical levels and numeric ranges. Refuses combinations no respondent could plausibly produce.
Interventional outputs
Interventional do(X) with downstream cascades through the fitted population. Move price; volume, share, segment mix shift the way your population would actually respond.
ACE & CATE estimates
Average causal effects across the population, plus conditional treatment effects across user-defined segments. With confidence intervals and positivity-gap reporting.
Methodology appendix
Stakeholder-ready document: assumptions, training data, scenario definitions, performance diagnostics, and known limits.
Validation diagnostics
Marginal MAE, multivariate MAE, variance preservation, K-anonymity-style novelty audit, infeasibility reports. Run on any customer dataset, anytime, by the Simulacra team.
Studio for the team. API for the stack. The same causal AI engine for both.
Most insights teams pilot the Synthetic Data Studio. Agencies, data teams, and agentic LLMs can control workflows in their stack through the headless API.