Research & systems studio

Systems, not prompts

Curate Labs builds knowledge-graph-based systems that encode how small businesses actually operate: finance, marketing, and compliance included.

We turn fragmented work into structured, traceable, and repeatable operations.

Our open-source work defines the primitives. HatTrickHQ applies them to deliver real-world operator leverage.

Explore research Explore HatTrickHQ

Infrastructure to application

A research-first stack that moves from explicit primitives to software operators can run.

Research

Model the real operating logic of a business as a system.

Open source

Publish reusable primitives others can inspect, fork, and extend.

HatTrickHQ

Turn those primitives into repeatable workflows for operators.

FinanceMarketingComplianceTraceability

Philosophy

How we think

We are a research and systems studio: explicit models, open-source primitives, and product work that operators can run without heroics.

How we think · 1

Knowledge graphs

over prompt chains

Represent entities and relationships explicitly so reasoning stays auditable—instead of hiding logic in prompt chains no one can replay or inspect later.

Research themes

How we think · 2

Systems

over outputs

We ship durable mechanisms, contracts, and workflows—not disposable generations that look good once and evaporate when context shifts.

Read perspectives

How we think · 3

Repeatability

over one-off wins

Real leverage is playbooks and instrumentation your team can run again next quarter—across finance, marketing, and compliance—without heroics.

View products

What We're Building

From research to products and open source

Commercial product direction alongside labs: open source repositories and research implementations you can read, run, and extend.

Open source · PyPI · alpha

GraphForge

Composable graph tooling for analysis, construction, and refinement—openCypher-compatible embedded engine with SQLite for research and investigative workflows.

View on PyPI

Open source · GitHub

InfoExtract

Open source utilities for extracting structured information from semi-structured sources—pipeline-friendly primitives for research and product experiments.

Open repository

Open source · PyPI · alpha

WACCY

Intelligent financial modeling for small businesses—extract and classify accounting data, build integrated models, and export institutional-style outputs (QuickBooks and EDGAR extensions on PyPI).

View on PyPI

Open source · GitHub

DocUnderstand

Document understanding for trustworthy AI—parsing, structure, and analysis primitives for long-form content in research and operator workflows.

Open repository

Research Themes

Current areas of inquiry

A few themes we return to often. See Research for the complete list.

Systems · structure

Explicit graphs and contracts

Model entities, relationships, and policies as first-class structure so automated reasoning stays inspectable—rather than burying logic in opaque prompt chains.

View research

Domains · workflows

Operator-ready AI

Finance, marketing, and compliance behave like interconnected systems in real businesses. We study how to encode that reality so tools fit how people actually work.

View research

Trust · documents

Documents, structure, and provenance

Long-form content needs parsing, segmentation, and traceable lineage when models consume it—primitives that hold up under audit and iteration.

View research

Latest Writing

Recent perspectives

Short-form arguments and operator notes from current work.

New perspectives will appear here soon.

Work on the business. Not just in it.

HatTrickHQ helps operators drive consistent wins across finance, marketing, and compliance—without adding complexity.

Explore HatTrickHQ