Senior Data Scientist

Andromeda builds personalised robot companions. We are pioneering the future of human-robot interaction where no playbook exists. We are backed by investors including San Francisco-based Forerunner Ventures, Rethink Impact and Main Sequence Ventures. We're working towards robots that feel genuinely alive – think Disney-level emotional connection. We have real customers, real end-users and have the runway to investigate and demonstrate technical breakthroughs.

What We're Building

We're creating a wellbeing insights platform. Our robots generate transcripts and perception data from resident interactions. The opportunity: turn this into insights that genuinely help aged care staff support their residents better. We need someone who has the judgement to hone in on what's useful for customers and end-users, build it, and prove it's valuable.

This is data science work with direct impact on vulnerable people's lives.

What You'll Own

You'll own the insights piece of this platform:

• Work with care managers, psychologists, and Human Robot Interaction (HRI) specialists to understand what questions matter (you're not the aged care expert, but you're excellent at translating between domains)
• Understand the extent and quality of data available
• Design experiments and build models across text, video and audio to provide actionable insights
• Ship insights that customers find genuinely valuable, measured by whether care managers actually use them to improve resident outcomes
• Build modularly so MVPs don't become technical nightmares six months later

The Working Reality

This is a 0-to-1 role, which means:

• Ambiguity is constant: Care managers want different things than executives. You'll need good judgement to deconflict contradictory requirements and align stakeholders, escalating to the Head of Machine Learning when genuinely stuck.
• Infrastructure evolves with you: We have searchable transcripts ready. Audio/video pipelines will come as you prove value. You'll flag what's missing but won't wait for perfect conditions. At the start, you'll need to build the queries and pipelines to feed your models, and the visualisation layer for their output.
• Quick wins matter: We need to demonstrate value to customers relatively quickly. Experiments can fail, but we need a nose for what will resonate and the tenacity to keep finding it.
• You ship MVPs: But you understand the tech debt you're incurring, document trade-offs, and build with enough modularity to avoid total rewrites later.

Who Does Well Here

You're someone who:

• Can build production-quality models but measures success by customer outcomes, not technical sophistication
• Can be a 'full stack' data scientist who can self-serve datasets, build models, test their efficacy, explain, and visualise their output
• Translates between technical and non-technical stakeholders naturally – you run workshops, synthesise contradictory inputs, and communicate findings clearly
• Has the humility to defer to domain experts on what matters whilst having the confidence to show them what the data reveals
• Thrives in early-stage chaos – ambiguous requirements energise rather than paralyse you
• Ships MVPs and knows the difference between "good enough for now" and "painting yourself into a corner"
• Doesn't give up when the first approach doesn't work

Day-to-Day

You'll spend time across:

• Customer conversations and workshops in care homes
• Building models, designing experiments, analysing data
• Collaborating with psychologists, HRI specialists, and university research partners
• Communicating insights and shaping product direction

The exact split is up to you – we care about what you deliver and when, not how you structure your time.

You'll Work With

• Head of Machine Learning (your manager, daily in-person collaboration)
• Chief of Product and Chief Commercial Officer
• Internal specialists and customer success teams
• External partners such as universities
• Care managers and aged care staff (who ultimately judge if your work is valuable)

Where This Goes

In 12-18 months, you'll lead data science as the wellbeing platform scales. We'll hire platform engineers and software engineers as needed. If you're capable and inclined, you can flex into other roles. We're a startup, but the primary path is building the data science function.

Requirements

• Bachelor's, Master's, or PhD in Computer Science, Data Science, Engineering, or related field
• 5+ years shipping data science products to real users (not just research or prototypes)
• Strong Python and modern software engineering practices
• Strong knowledge of SQL and ability to work with structured and unstructured data
• Demonstrated ability to navigate ambiguous, contradictory requirements and drive to valuable outcomes
• Excellent communication – can explain technical work to non-technical stakeholders and facilitate alignment
• Melbourne-based, five days on-site

Valued Experience

• Consulting background working with diverse customers and stakeholder landscapes
• Healthcare or aged care projects
• Full-stack development capability
• Experience with personally identifiable information and ethical data governance
• Building 0-to-1 products where you helped define what "valuable" meant alongside customers