The field-sales app 100+ companies run their business on. I redesigned it end to end, built a robust design system an AI agent can build screens from, and made it configurable enough that one product fits four very different industries.
The app already ran real business for 100+ companies. But two things were quietly holding it back, and they fed each other. The design was stuck in the past with nothing underneath to fix it, and one product was trying to serve far too many kinds of people at once.
The first job was to make it feel like one modern product. I rebuilt the interface on a real design system, one set of components and tokens, so every screen finally spoke the same language. Then I made that language work for the field: fast, offline, one-handed, almost no typing, always pointing at the next thing to do.






The redesign was the surface. Underneath, the app had to solve the messy, real problems a sales org lives with every day, for the rep in the field and the manager watching the whole team. These are the ones that mattered most.
It runs the daily field execution for 100+ enterprises across 4 industries, adopted by the users who are hardest to win over.
What stayed with me is that the app never won people over by looking new. It won because it hid its own complexity, one product bending into four different businesses while every rep still felt it was built only for them.