Designing a shared workspace for data scientists

The challenge

Data scientists moved between disconnected tools and resources to create and manage experiments. The product needed to bring that work together without hiding the different stages and actions involved.

I focused on three connected decisions:

1. Map work across disconnected tools.
2. Bring projects and resources into one workspace.
3. Confirm important actions clearly.

1. Mapping work across disconnected tools

I partnered with data scientists to map their end-to-end journey and identify repeated, time-consuming steps. I contributed the interface perspective while we tested assumptions and agreed the product structure together.

Key decisionUnderstand the workflow before defining the product

The map exposed repeated tasks and gaps between tools.

2. Bringing work into one workspace

I then led the UI, motion, and illustration work, partnering with product owners, business analysts, and experience designers to turn the agreed journey into the interface.

I used the mapped stages to organise projects, resources, and experiments around the way data scientists worked, rather than around the separate tools they previously needed.

Key decisionOrganise the product around the workflow

The interface gives teams a consistent way to create and manage experiments.

3. Confirming important actions clearly

I used focused motion to confirm important actions such as creating or deleting a project.

Key decisionReserve motion for meaningful feedback

Motion gives immediate feedback for project actions.

What changed

• Disconnected tools were represented through one product structure.
• Repeated project tasks gained consistent patterns.
• Critical actions received immediate feedback.
• The team gained a shared view of the workflow.

Instead of moving between disconnected tools, data scientists could manage projects, resources, and experiments in one workspace, with consistent feedback for critical actions.