About

I'm a forward deployed data engineer from Sydney, Australia. I embed with client teams to design, build, and productionise data platforms end to end, from discovery and architecture through to the systems that run in production.

What I do

Over the past few years I've built platforms across Databricks, Snowflake, BigQuery, and MotherDuck/DuckDB, on GCP, Azure, and AWS. I've moved clients onto Databricks and off it, onto Snowflake and off it; the right platform depends on the problem you're solving. Having built on the alternatives, Databricks is the one I've chosen to go deep on - the platform I'd bet on for most teams, and where I'm deliberately concentrating my craft.

The work runs the full width of a platform. On ingestion, that means event-driven architectures on Pub/Sub and Kafka, custom Python frameworks built with dlt, and CDC pipelines, with managed connectors like Fivetran and Polytomic where they earn their place. One system I built ingests more than 300 million rows a day from a fleet of residential IoT devices.

On transformation, it's dbt at enterprise scale, authoring over 2000+ models in production, alongside dimensional modelling, and semantic layers. Underneath sits the DataOps that keeps a platform maintainable well beyond a single engineer - Dagster and Airflow, Terraform-driven infrastructure, CI/CD, and real test coverage. Increasingly the work reaches into governance and AI too, from Unity Catalog and privacy-first access control to RAG pipelines, Genie Spaces, Agent Bricks and agentic tooling like Claude Code and MCP applied to data and analytics engineering.

Where I've done it

The settings have been varied, and that variety is the point. I started as a data engineer at Mantel Group, working up from associate to senior across enterprise consulting. There I set up the foundational architecture for migrating one of the country's largest energy retailers off a legacy Databricks estate onto Unity Catalog–enabled workspaces, building the config-driven Terraform and reusable Azure DevOps pipelines the cutover ran on. Other work there ran the gamut: 800+ dbt models on a national retailer's move off Teradata to GCP, a consolidated risk model on dbt, orchestrated on Airflow, that merged three sources and around two billion rows for organisation-wide exposure assessment, and a GenAI pipeline on Vertex AI that turned around 20,000 free-text survey responses into structured insight for leadership.

From there I moved into founding and forward deployed roles. I was the founding data engineer at a Series A energy startup, standing up an event-driven platform on GCP to run a virtual power plant in real time. Through my consultancy I built a config-driven ingestion framework that cut one client's costs by roughly 95%, and more recently, as a forward deployed engineer, I grew a single strategic account to $2M in eight months by owning its data platform end to end.

Teaching

I also teach. I've delivered data and analytics engineering courses to more than 100 engineers, and launched an AI engineering curriculum covering agent harnesses and MCP for production analytics work.

Background

I came to data the long way around: a PhD in chemical engineering at UNSW, then data science, and finally the engineering work I enjoy most. I'm certified across Databricks (Professional Data Engineer), GCP (Professional Data Engineer), dbt (Cloud Architect), Azure, and HashiCorp. These days I take on platform builds, migrations, and enablement through NimbleStax, my consultancy. The full timeline is on my CV.

Let's connect

I'm always open to conversations about platform builds, migrations, and enablement. The fastest way to reach me is by email.