The Problem
Tight deadlines. A moving target. Real stakes.
Every year, ahead of a major international product launch, the client needs a fully operational prototype that replicates a mobile OS for internal tech support use. Staff use it in place of the physical device to walk customers through issues in real time.
The catch: the OS is still in beta during development, which means the target is moving the entire time. This requires tight communication and precise execution to maintain quality across every iteration.
Challenges
Three things that made this hard.
- Real-Time Adjustments Continuously adapted to changes in a Beta operating system, requiring rapid iteration while maintaining design quality.
- High-Volume Design Work Managed a large volume of design tasks in a fast-paced, high-pressure environment, balancing efficiency with precision to meet deadlines.
- Four Languages Produced screen sets in four languages across both years. I don't speak any of them, and didn't need to — the job was matching localized screens to spec exactly, which is a care problem rather than a language one.
Year by Year
Two years. Two very different roles.
2025
Came back as a contract lead designer, on a team of five designers and two other design leads. Everyone except me was new. I onboarded and trained all of them, turned my own workflow from the previous year into the documented team standard, ran training workshops, and wrote the reference document people worked from. Four team members were falling behind, so I ran one-on-one sessions with each of them. Daily check-ins with everyone else.
The part I’d lead with: I owned the integration of every designer’s individually prototyped app into a single working prototype of the entire device. That had never been delivered complete in any prior year. It was delivered complete this time — the PM told me so directly. The off-site team lead taught me the basics of the integration tool over a few meetings, and I worked the rest of it out and executed it myself.
Midway through, two of the other design leads dropped off the project. I absorbed their responsibilities and became the primary point of contact with the client, which the other design leads hadn’t been doing. I was still producing screens alongside the team the whole time.
2024
An off-site team handled the prototyping itself. My job was producing and preparing the screen assets they built from: capturing screenshots on device across the OS apps, importing them to Photoshop, stitching scrolling screens together, and resizing everything to exact spec. Hundreds of screens a day.
It was tedious by nature, so early on I went looking for a better way. I tried to write custom automation with Claude Code first and couldn’t get it working at the time. Instead I sourced existing Photoshop actions and scripts and chained them into a pipeline covering import, resize, and export. A manual step stayed in for layout. It was assembled from existing tooling rather than written from scratch, but it held up, and the output was more consistent for it.
The clearest signal it was working showed up on the back end: my screens usually needed one or two rounds of revision from the off-site approval team. Three or more was common for others, and some hit eight. The room that created let me take on overflow from three other designers while keeping my own pace. I volunteered for it first, then became the person it went to by default. At the end of the project the leads asked me back the following year as a design lead, and to bring the workflow with me.
Results
Can’t show it. Here’s what it does.
On the ground, it does what it was built to do: support staff can walk a customer through a device issue without needing the physical hardware in front of them. That’s the outcome.
What I’d do differently: The workflow handoff in 2025 went well overall, but I underestimated the onboarding curve. Enough of them got confused early on that I ended up doing the coaching after the fact instead of building it into the handoff. I saw it coming and didn’t do enough about it. Next time I’d build a proper onboarding session before anyone touches the workflow — not a quick demo, an actual walkthrough with time to ask questions.
Takeaways
What this taught me.
Anticipating Mistakes
You can't assume people know how or when to do something. Being a leader means spotting where things are likely to go wrong and preemptively guiding the team.
Thriving in Fast-Paced Environments
Both years confirmed something I already suspected: I do my best work when the pace is high and the stakes are real. Tight timelines don't drain me. They focus me.
Resilience Under Pressure
Things will inevitably go wrong. When two design leads left mid-project, I was able to regroup, refocus, and keep the project moving forward without losing momentum.
Get in touch
This project is under NDA, but I’m happy to walk through it in more detail. Get in touch.