About us

    We turn opaque systems into explainable ones.

    Complex systems become opaque faster than teams can document or reason about them. We build deterministic, queryable models of code and data that engineers and AI systems can both rely on.

    Sydney, Australia · Working with teams across AU/APAC and remotely

    Our story

    Born in financial markets.

    George Jayaratnam spent more than 25 years in financial markets in Sydney, building risk models and the data pipelines behind them. One problem kept resurfacing: complex systems become opaque faster than teams can document or reason about them.

    That led us to knowledge graphs and GraphRAG workflows: building deterministic, queryable models of code and data that can be used by both engineers and AI systems.

    Today, working with teams across AU/APAC and remotely, we help organisations build Codebase Knowledge Graphs — enabling safer refactoring, better system understanding, and more reliable AI-assisted development.

    Areas we specialise in

    • Financial risk modelling and decision systems: risk optimisation models
    • Enterprise data platforms and system migrations: data lineage tracing
    • Large-scale codebase analysis with graph models, in the open-source tools we publish on GitHub
    • GraphRAG workflows that constrain and validate LLM output
    Our goal
    To turn complex, opaque systems into structured, explainable and queryable assets.
    How we work

    Four principles behind every graph.

    01

    Structure first

    We extract what can be known deterministically before any model gets involved.

    02

    Grounded in sources

    Every concept in the graph traces back to code, schema or a cited reference.

    03

    Built iteratively

    The graph schema evolves with your questions and your codebase.

    04

    Designed to act

    Graphs that drive decisions and checks — not documentation that sits on a shelf.

    Get started

    Turn your codebase into a queryable, explainable system.

    Request a discovery call. We’ll talk through your system, where it hurts, and whether a Codebase Knowledge Graph is the right fit.