What We Do

We help you transform raw data into actionable insights and deliver scalable AI/ML solutions tailored to your business needs.

Our Process

We focus on understanding your business, building practical solutions, and delivering results through a structured approach:

What This Looks Like in Practice

Each project follows our structured process, but outcomes, timescales, and commitments differ depending on business goals. Here are examples of practical results we've delivered for clients:

• Enabled semantic search across documents for medical, legal, and insurance domains.
• Custom classification model to identify suspicious transactions in a fintech context.
• Custom retrieval-augmented generation (RAG) system to support market research.
• AI-based image analysis in a medical diagnostics workflow to assist radiologists.
• Advised on self-hosting strategies for privacy-aware AI deployments.
• Feasibility studies for AI tooling in businesses with low data maturity.
• 12-week, standalone proof-of-concepts to gauge expected ROI.

How might your project look? Contact us to explore.

Technical Articles

Structured knowledge from client work, explorations, and deployment challenges.

• LLM Training Fundamentals
• Platforms vs Pipelines
• RAG Strategy and Tooling
• Why are Vector Databases Difficult?

All articles →

Projects and Open Source

Tools and products built from client work and independent research.

⚘

Marigold

Typed inference pipelines over privately hosted open-weight models. Define your task, build your pipeline, measure and improve.

Learn more →

A multilingual children's story generator. Choose a language, a theme, and a cast of characters -- Popstory writes the story. Built with and for younger readers.

Visit Popstory →

runfox

Declarative workflow execution engine. Define dependency graphs in YAML; runfox resolves and advances them.

PyPI →

json-logic-path

JSONLogic extended with JSONPath variable resolution. Compose conditional logic as data structures.

PyPI →

dynawrap

A thin, typed wrapper over the AWS DynamoDB client. Dataclass-based item definitions with clean get, put, and query patterns.

PyPI →