Software engineer · Prague

Economist by training. Engineer by choice.

I studied financial economics in Toronto, then software engineering in Prague. Now I build iOS and Android apps, with AI doing most of the typing and me deciding what's worth building and how it should work.

Portrait of Karolína Droscová

About me

Economics was the plan. Then I got curious about the code behind the tools I was using—and wanted to learn to build them myself. These days, I mostly build mobile apps. My economics background still shapes how I work: untangling complex problems for clients, figuring out what’s worth building, and seeing it through.

Read the full story →
  • BSc Financial EconomicsUniversity of Toronto · GPA 3.98
  • BSc Software EngineeringFIT CTU, Prague
  • 70k+ monthly usersSlevomat app, iOS & Android
  • Dean's List2020 · 2021 · 2022

How I think

Code is cheap. Thinking isn't.

With AI writing more and more of it, the valuable work is everything around the code: owning the product, being a real partner to the client and getting the architecture right.

I do miss writing code sometimes, though.

Every feature is an investment.

I look at a feature the way I learned to look at an investment: what it costs, who it's for and what it returns. Sometimes the best engineering decision is not to build it at all.

Even if it means I don't get paid to build it.

My niche isn't my limit.

I understand how software is built, so I can build in languages and frameworks I haven’t used before. AI helps fill in the syntax, and I learn the rest as I go.

This website? The first one I’ve ever built.

What I build

My work, in practice.

Mobile apps

I build native iOS and Android apps, sharing as much logic as possible with Kotlin Multiplatform. I help shape features, keep the architecture clean, and see the work through testing and release.

  • iOS
  • Android
  • Kotlin Multiplatform
  • Swift / SwiftUI / UIKit
  • Kotlin / Compose
  • App architecture
  • Testing
  • CI/CD
See it at Slevomat →

AI development workflows

I build tools that help AI agents work with a codebase and check their results. That includes reusable rules and skills, automated app testing, and build workflows that support working on several tasks at once.

  • Claude Code
  • Codex
  • MCP servers
  • Rules & skills
  • Agentic Review & QA
  • Parallel workflows
  • CI/CD

Selected work

Products I've helped build.

Client project at MeguMethod

Slevomat

70k+ monthly users

I build features across Slevomat’s iOS and Android apps, move application logic into shared Kotlin code, and improve automated testing and release workflows.

  • iOS
  • Android
  • Kotlin Multiplatform
  • Swift
  • SwiftUI
  • Kotlin
  • Web views

Client project at MeguMethod

Záchranka for Apple Watch

5 country apps · one shared watchOS core

Taking the lead on Záchranka’s Apple Watch companion from early discussions to working builds for five country apps. A chance to shape the product, build its foundation and see it through.

  • watchOS
  • iOS integration
  • Swift
  • SwiftUI
  • WatchConnectivity
  • WidgetKit

Client project at MeguMethod

Fintech app with virtual payment cards

10k+ users

From native iOS to Kotlin Multiplatform: learning a new language while delivering features in my first job after school, with teams across three companies building the product together.

  • iOS
  • Android
  • Kotlin Multiplatform
  • Swift
  • SwiftUI

I'll tell you if it's worth building. Then I'll build it.

Whether it's a role, a product you want to build, or just a conversation about KMP or AI workflows, I'd love to hear from you.