Job Description

We are a global research and advisory firm that focuses on the world's most consequential geopolitical and strategic matters. Our work informs the decisions of governments, institutions, and enterprises.

n

We hold ourselves to unusually high standards. Our research is defined by empirical rigour and integrity. Our writing by clarity and authority. Our advisory work by genuine mastery of subject matter.

n

We are a flat, expert-heavy team. Everyone operates at a high level, takes ownership, and is trusted to exercise judgment.

n

The Role

n

You'll work as part of a team designing and building the firm's systems — the channels clients use, the research and authoring tools our analysts work in, the data lakehouse and analytics layer underneath, with AI used throughout.

n

The job runs from strategy to code: you'll help work out what the firm needs, design it, then build it — working directly with the CTO, the rest of the human tech team, and an ensemble of AI agents.

n

n
- Languages: Python and TypeScript (front-end), occasionally Rust.
n
- Cloud: AWS — many services, IaC and containers throughout.
n
- Data: Aurora/Postgres, Iceberg, and a range of analytical and graph databases and caches.
n
- CI/CD: GitHub Enterprise and Actions, with security embedded in the pipeline — reviewed, gated change as the default (DevSecOps).
n
- AI in the delivery loop, with agents and humans in a hybrid team.
n

n

You

n

We're looking for someone with three to five years of experience, with strong software engineering fundamentals. You should have broad skills across networks, operating systems, cloud (AWS), databases, and data science.

n

Nobody's an expert in everything — we care more about versatility and how rapid you pick things up. But we expect that:

n

n




- You've built something hard. The work you're proudest of took real engineering — a data platform, a distributed system, an ML pipeline, an inference-backed feature — and you can take us through what made it difficult and how you solved it.
n
- You move between languages easily. You reach for whichever one fits and don't get precious about it.
n
- You understand data, analytics, and ML. You've modelled data, built pipelines that held up in production, and worked with ML past the coursework stage. You've taken models from a notebook into production yourself.
n
- You know what's under the application. E.g. networking, operating systems, cloud primitives, databases. When something breaks a couple of layers down, you can find and fix it.
n
- You learn fast and get things done — you'd rather ship something and iterate than wait to be handed a spec.
n
- You use AI well. Claude Code and similar tools are part of how you build, and you can help build the infrastructure AI workloads run on.
n
- You're self‐directed and honest with yourself — you find answers, and you know when to go deep and when to ask.
n
- You take probity and confidentiality seriously.
n
- Three years' professional software engineering minimum; the role is pitched around five.
n
- Production Python and TypeScript — systems you shipped and kept running, not one‐off scripts.
n
- A system you owned end to end, and can walk through in depth.
n
- Real data‐engineering experience — pipelines, modelling, serving analytical workloads.
n
- AWS through IaC — Terraform and containers, with the console as the exception.
n
- DevSecOps fundamentals — pipeline security, least privilege, secrets hygiene.
n

n

Desirable: a degree in Computer Science and/or Data Science.

📌 Software Engineer (Victoria)
🏢 Garnaut Global
📍 Victoria