Unlock the Value of Dark Data

Dark data is unstructured data such as text and images.
Lattice turns dark data into structured data
with human-caliber quality at machine-caliber scale.

Dark Data


Entities are people, places, things, customers, companies, law firms, doctors, stocks, genes, diseases, chiropractors, etc. Lattice extracts and integrates information around entities from all data sources, be they mentioned as Mr Smith in voice transcripts, the senator in notes, or 00248 in tables.


Relationships between entities may be spouses, family, employment, clients, partners, owners, friends, subsidiaries, etc. They are often buried in natural language and semi-structured text. Lattice uses advanced machine learning to unearth such data, regardless of linguistic variations and domain idiosyncracies.


Events are a combination of what, who, when, and where. They may be personal life events, corporate events, financial transactions, calls and emails, meetings, announcements, natural disasters, etc. Lattice unifies diverse data sources to detect events and reconstruct timelines.

Structured Data


Stanford DeepDive

Lattice was born out of the Stanford research project DeepDive, a framework for statistical inference. We are the researchers and creators behind DeepDive, and now we build on top of DeepDive to create products for industry and government. Our mission is to unlock the value in dark data for critical real-world problems.

Statistical Inference

DeepDive is a programming and execution framework for statistical inference, which allows us to solve data cleaning, extraction, and integration problems jointly. We model the known as features and the unknown as random variables connected in a factor graph.

Impactful Applications

DeepDive has been successfully applied to a diverse set of projects.

Bootstrapped Learning

Unlike traditional machine learning, we do not require laborious manual annotations. Rather, we take advantage of domain knowledge and existing structured data to bootstrap learning via distant supervision. We solve data problems with data.

Machine Scale

We continuously push the envelope on machine learning speed and scale with our bleeding-edge systems research. For years, we have been building systems and applications that involve billions of webpages, thousands of machines, and terabytes of data.

Human Quality

Data quality is in the DNA of Lattice. Our goal is not just to match human-level quality, but also to do so at unprecedented speed and scale. We build systems that win competitions and outperform expert readers.

You may reach us at


801 El Camino Real, Menlo Park, CA 94025