Anduril Australia

Lattice

Designing command and control dashboards for one of Australia's most significant autonomous defence programmes

This project was conducted under a non-disclosure agreement. What follows describes my design process and approach using only publicly available information about Anduril and the Lattice platform. Specific details about features, data, and design outputs have been omitted.

EmployerHide and Seek Digital

SectorDefence, Autonomous systems, AI/ML

My roleSenior Product Designer

Team3 (Project lead, 2 developers)

Time3 months

I was the sole designer on a focused team embedded within Anduril's engineering organisation. I was responsible for the end-to-end design of two new dashboard modules within Anduril's Lattice command and control platform, from discovery through to validated designs ready for development.

I worked within Anduril's existing design system, collaborating with their US-based design team, Australian engineering leads, and programme stakeholders.

A$1.7B

Ghost Shark programme investment

Dashboard modules delivered

3 months

Discovery to validated designs

The problem

Anduril's Lattice is the AI-powered software platform that provides command and control across their family of autonomous systems. In Australia, their highest-profile programme is Ghost Shark (also known as Dive-XL), an extra-large autonomous underwater vehicle developed with the Royal Australian Navy, representing a A$1.7 billion investment in sovereign maritime capability.

Autonomous underwater assets produce vast amounts of operational data. When our engagement began, the existing approach to monitoring these assets within Lattice wasn't purpose-built for operational decision-making. Operators were presented with large volumes of raw data, most not relevant to their immediate tasks, and there was no consolidated view for managing multiple assets simultaneously.

The brief: Design two new dashboard modules within Lattice, one for single-asset monitoring and one for multi-asset fleet management, giving operators the situational awareness to make quick, informed decisions during mission planning and execution.

Research and discovery

I facilitated three workshops in the opening weeks, each targeting a different dimension of the problem: understanding operators and their decision-making patterns, systematically prioritising which data points mattered most for each dashboard context, and mapping the technical landscape to surface constraints and data availability.

Two insights shaped the direction:

Single-asset and multi-asset monitoring required separate modules. The decision-making patterns were different enough that trying to serve both in a single interface would have created the same data overload the engagement was trying to solve.
Data prioritisation was the critical design input. Without a structured exercise to separate what operators needed at a glance from what could be accessed on demand, the dashboards would have reproduced the existing problem in a new layout.

Discovery workshop outputs in Miro — Sample of workshop outputs: job stories mapped across operator types (left) and data prioritisation using a MoSCoW framework (right).