Machine Learning Engineer / Applied Scientist

The role 👤

We're looking for someone who thinks like a scientist and wants to learn to build, automate, and deploy like an engineer.

You'll join our Data & AI team: a small, diverse group that builds and operates data products across the business. Fraud detection, customer service automation, intelligent query routing, unstructured data analysis, that kind of thing. Every person on the team brings a different background. What we all share is that we own our work from the initial idea all the way into production.

What we're adding with this role is a new lens: quantitative research and behavioural science. Your understanding of how and why people behave the way they do will strengthen the entire team's ability to work on customer intelligence problems: identifying common attributes of our customers, what drives their behaviours, and how to better engage with them.

On a normal day, you could be:

• Applying causal inference and inferential statistics to figure out what's actually driving customer behaviour vs. what's just correlated with it.
• Training, evaluating, and automating models that will integrate directly into our systems as live features.
• Translating model outputs into strategic recommendations, ensuring that "Customer Intelligence" leads to measurable improvements in experience.
• Taking existing customer segmentation research and automating it into a scheduled pipeline that can be acted on and used beyond reporting.
• Partnering with our Customer Experience and Analytics teams to build the automated models and self-serve tooling that will elevate how the entire company understands and serves our customers.
• Growing your engineering skills. You bring at least a baseline of clean Python, Git, and SQL; we’ll teach you our engineering standards and best practices to ensure your work scales seamlessly.

Where you'll work: Melbourne (hybrid - in office when it matters, WFH when you need focus time)

You should apply if… ↪️

You have:

• Quantitative research chops. A postgraduate degree (or equivalent through industry experience) in a field where causal inference is core: econometrics, biostatistics, quantitative psychology, computational social science, epidemiology, or similar. You know when a causal claim is defensible and you choose methods based on the question, not the tool.
• 2-5 years in the industry applying quantitative methods, ideally in a customer-facing or product context. You've produced analysis that actually changed a decision.
• Strong inferential statistics. Hypothesis testing, confidence intervals, regression, causal inference methods (diff-in-diff, propensity score matching, instrumental variables). Plus solid ML fundamentals: clustering, classification, predictive modelling and the judgment to know when a simple regression beats a deep learning model.
• Python and SQL proficiency. Experience with scikit-learn, pytorch, XGBoost, or similar. You write code someone else can read and maintain.
• Engineering willingness. You don't need to be a software engineer today, but you want to become one. You're drawn to teams that build and automate, not teams that just report. You see version control, testing, and code review as growth, not overhead.
• Commercial instinct. You think in terms of decisions, not just findings. "So what?" is your favourite question. The complexities of a model are less important to you than how it affects customers.
• Communication skills. You can defend your methodology to a statistician and clearly explain the implications to a non-technical product manager. No jargon fog.

Bonus points for:

• Behavioural science background. Psychology, sociology, anthropology, behavioural economics. Intuition about why people do what they do, not just a model that predicts what.
• CX domain knowledge. NPS, CSAT, customer lifecycle, Voice of Customer data.
• Experience with data technology. Big data analysis, data pipelines, orchestration, ML model packaging, and ML model serving.

You'll thrive here if you:

• Get excited by "why is this happening?" more than "what happened?"
• Want your segmentation model to actually do something in production.
• Like working in a small team where everyone knows everyone's work.
• Think the best analysis is 30% technique, 70% asking the right question.

This probably isn't for you if:

• You want to work in isolation and hand off a report.
• Automating and deploying your own work sounds like someone else's job.
• You need every research question perfectly scoped before you start.

How we work

💪 Our tech stack:

• Data warehouse: BigQuery
• Transformation: dbt
• Orchestration & Infrastructure: Dagster, Docker
• BI: Metabase
• Languages: Python, SQL
• Collaboration: Slack, Notion

Team structure:

• You'll report to the Technical Director of Data & AI.
• You'll work alongside ML Engineers and Data Engineers.
• You'll partner regularly with the Customer Experience team, product managers, and other teams across the business.

How we engage with other teams: We work in partnership with teams across Up on defined projects each season. Teams come to us with problems, we scope them together, and we build solutions that last. It’s a two-way street as well - we actively search for systemic friction to remedy, as well as commercial opportunities to exploit in our products. We collaborate, we build, and we deploy.

FAQs 💭

Do I need to be a software engineer? Not necessarily. However, you should have an active interest in becoming one. We'll invest in getting you there; our team is built for this.

Will I only work on CX projects? No. You’ll undertake customer intelligence work, but you're a full member of the team. You'll contribute to projects across the business.

Do I need a PhD? Not necessarily. If you've built equivalent depth through industry experience, that counts. We care about what you can do, not what's on your wall.

What if I'm more of an applied scientist than an engineer? Great. That's what we're looking for. We're hiring for your research brain, and we can support your engineering growth.

Is this role remote? Hybrid. Melbourne-based. We value getting together but we optimise for output, not chair time.

Our salaries

As part of our commitment to a fairer workplace we internally publish salary bands. All of our offers will be within these bands, and the final offer will depend on where within the level our assessment places you.

Our salary bands are as follows:

• IC4 (mid-level) pays from $125,000 to $145,000
• IC5 (senior-level) pays from $150,000 to $170,000

All listed salaries are base pay, exclusive of superannuation and potential bonus equity (dependent on minimum tenure and group performance).

If you aren’t sure if this is the right role for you, or you’d like to have a chat with someone who works here first to ask any questions you might have, send us an email at careers@ferocia.com.au and we’ll pair you up for a chat with an engineer or engineering leader.