DataCebo
DataCebo is an AI company focused on solving data-availability problems with AI-generated synthetic data. Its Synthetic Data Vault (SDV) software builds generative relational models of enterprise databases so organizations can generate synthetic data for testing, AI training and evaluation, scenario simulation, and data sharing while keeping data in their secure environment.
At a Glance
- Global 2000 and large enterprises
- Financial services and banking
- Insurance
- Healthcare
- +4 more
AI Tools by DataCebo
(2)
SDV (Synthetic Data Vault)
Python Library for Synthetic Data
CTGAN
Python Synthetic Tabular Data Generator
Discussions
No discussions yet
Be the first to start a discussion about DataCebo
Latest News
Published 'Critical capabilities for AI-generated synthetic test data'.
Announced SDV 2.0 and generative relational models for enterprise relational data.
Published 'Enterprise AI should train where enterprise data lives'.
Launched differentially private synthesizers in SDV.
Products & Services
Commercial, enterprise software for building generative relational models from complex databases and generating secure synthetic databases at scale inside the customer's environment.
Free, source-available SDV platform for single-table, multi-table relational, and sequential/time-series synthetic data generation.
Major SDV release that automates building a reusable generative relational model of an enterprise database from representative data.
Open-source platform/library for evaluating synthetic-data quality.
Market Position
DataCebo positions SDV as an enterprise-owned generative relational-model stack purpose-built for structured, interconnected enterprise databases, rather than a general-purpose language model or a conventional masking/ETL test-data tool. Relevant alternatives and competitors include synthetic-data vendors such as Gretel, Mostly AI, Tonic.ai, Hazy, and Synthesized, while DataCebo differentiates with multi-table relational modeling, business-rule preservation, on-premises deployment, and its open-core/source-available SDV ecosystem.
Leadership
Founders
Kalyan Veeramachaneni
Co-founder and CEO; Principal Research Scientist at MIT and founder/leader of MIT's Data to AI Lab. He has spent more than 15 years researching AI for relational and tabular data and previously founded the Data to AI Lab at MIT in 2015.
Neha Patki
Co-founder and chief product leader (described as VP of Product at launch and CPO in later coverage). She invented the first generative AI for relational/tabular datasets and previously worked as a product manager at Google, managing creator-analytics features at YouTube; she studied computer science and artificial intelligence at MIT.
Executive Team
Kalyan Veeramachaneni
Co-founder and CEO
Principal Research Scientist at MIT and leader of the MIT Data to AI Lab; researcher and entrepreneur focused on relational/tabular AI and synthetic data.
Neha Patki
Co-founder and Chief Product Officer
Product leader who previously managed creator analytics at YouTube/Google; co-created SDV and invented generative AI for relational tabular datasets.
Board of Directors
Founding Story
The Synthetic Data Vault project was first created at MIT's Data to AI Lab in 2016. After four years of research and enterprise traction, DataCebo was created in 2020 to grow the project. Kalyan Veeramachaneni and Neha Patki originally built SDV to remove barriers to data access and availability and accelerate software development; they later commercialized the technology for enterprises that needed realistic, private, application-valid data.
Business Model
Revenue Model
A free community/source-available product builds adoption, while SDV Enterprise is sold as per-user software with a base subscription plus usage-based pricing; optional SDV Bundles are sold as add-ons and synthetic-data generation is pay-as-you-go.
Pricing Tiers
Free platform with limited commercial use under the Business Source License; community/forum support.
Enterprise base plan with 12+ models, 10+ data types, and dedicated support; AI Connectors available as an add-on.
Optional add-ons including CAG, XSynthesizers, Targeted Sampling, Differential Privacy, and AI Connectors.
Customers pay for the synthetic data they create and can set a spending cap.
Target Markets
- Global 2000 and large enterprises
- Financial services and banking
- Insurance
- Healthcare
- Software and application development
- AI/ML teams
- Software, application, agentic, regression, performance, and end-to-end testing
- Training and evaluating AI models, copilots, and agents
- Scenario simulation and what-if analysis
- Fraud detection and rare-outcome modeling
- Market research and synthetic survey/behavioral data
- Data sharing and collaboration without distributing sensitive records
- ING Belgium
- MAPFRE
- Spar Nord