SwiftBoost is my own product — PC optimization software for gamers, with one-click optimization profiles that reduce latency and boost frame rates. I designed it in Figma and shipped it as a working application, end to end. It's on this page for a reason: when there's no client to satisfy, what you build shows exactly what you'd do with full freedom — and this is that.
Context
Gamers obsess over frame rates, but most PC optimization is either snake oil or a maze of registry tweaks scattered across forum threads. The people who'd benefit most — players on mid-range hardware trying to squeeze out playable FPS — are exactly the people least equipped to safely change system settings by hand.
SwiftBoost's premise: collapse all of that into optimization profiles a user applies with one click. No forum spelunking, no risk of breaking your machine following a five-year-old Reddit post.
Constraints
- Trust is the whole product. Software that touches system settings has to feel safe. A janky UI reads as malware; the design had to look like something you'd let near your OS.
- The audience is impatient by definition. Gamers chasing lower latency won't tolerate a sluggish settings app. The tool itself had to be fast.
- One person, full pipeline. I was the designer and the developer — every screen went from Figma to code without a handoff.
The build
1. Design-first in Figma, then build
I designed the full interface in Figma before writing application code. For a product where trust is the feature, this mattered: I could iterate on visual credibility cheaply — spacing, hierarchy, how an "optimize" action communicates what it's about to do — before committing anything to code. Being a one-person pipeline meant zero drift between mockup and shipped UI.
2. React + TypeScript + Electron
The interface is React with TypeScript, packaged with Electron so it runs as a desktop app with access to the system it's optimizing. TypeScript earned its keep here — when your app changes system behavior, "it compiled so the data shapes are right" is real risk reduction, not ceremony. And building on web tech meant the same component work powering the app also powers the product site at swiftboost.zadmin.dev.
3. Profiles, not switches
The core UX decision: users pick a profile, not a hundred toggles. Bundling tweaks into curated one-click profiles keeps users out of dangerous territory and makes results repeatable — which is also what makes a claim like "average FPS increase" measurable instead of anecdotal.
Results
The headline: +40% average FPS increase for users applying SwiftBoost's optimization profiles.
What's verifiable from the work itself: a designed-in-Figma, built-in-React/TypeScript desktop application, live at swiftboost.zadmin.dev, that packages risky manual system tweaks into safe one-click profiles.
Want something like this?
SwiftBoost is what it looks like when I take a product from blank Figma canvas to shipped software, solo. If you've got an app idea — internal tool, SaaS, desktop utility — see my web apps service, check pricing (custom builds start from $6,500), or tell me about it.