Open Source Products

Tools we build. Problems we solve.

Alongside consulting services, we develop open source frameworks that address fundamental challenges in data architecture and AI coordination. Enterprise implementation and consulting available for both.

Measure the cost of AI. Automate what's proven.

DIAL is a coordination framework for AI and human specialists making decisions together within state machines. It answers a question most AI deployments ignore:

Given any task modeled as a state machine, how do you know, in dollars, time, and quality, exactly what it would cost to turn that task over to a minimally competent AI decision-maker?

Human Primacy

AI has no role by default. Humans remain authoritative because they possess contextual knowledge inaccessible to AI: lifetime experience, institutional knowledge, and real-time sensory input.

Progressive Collapse

As AI alignment proves itself through repeated measurement, the system gradually delegates decisions to the most trusted specialist, with automatic reversion if alignment degrades.

Empirical Trust

Trust develops through demonstrated alignment, not assumptions. Every decision point generates dollar-precise cost data, alignment rates, and latency measurements.

$0.003 per AI decision

94.2% alignment rate

~200ms decision latency

Learn More About DIAL View on GitHub

Describe your data in a markdown-like format. Run your entire data strategy on it.

Eloquent is a unified semantic data modeling language designed for the age of LLMs. Write a simple, human-readable data dictionary using natural language relationships, and Eloquent generates your entire data architecture: physical schemas, logical views, knowledge graphs, data catalogs, SQL tests, and LLM-ready RAG context.

- Customer has a Customer Name
- Customer Name is a String
- Order has a Customer
- Order has a Price
- Price is a Number

# Eloquent automatically derives:
> Customer has an Order Count
> Customer has an Avg Order Price
> Customer has a Min Order Price
> Customer has a Max Order Price

Natural Language Modeling

Define entities, attributes, and relationships using markdown-like syntax that reads like English. No YAML, no JSON schemas. Just describe what your data means.

Full-Stack Generation

From a single model definition, generate 12+ artifacts: physical schemas, views, ERD diagrams, data catalogs, SQL tests, observability scripts, and RAG-compatible context for LLMs.

LLM-Native Architecture

Models are designed from the ground up for natural language querying. Generate NL-to-SQL training pairs, vector-database-ready representations, and golden test suites.

Learn More About Eloquent View on GitHub

Need help implementing?

Both products are open source. Enterprise implementation, custom integration, training, and ongoing consulting are available.

Schedule a Conversation