Tapnite
Tapnite
Senior Game Developer
I make games that feel good to play, and build the systems behind that feeling.
6+ years shipping mobile games · EU-based · Tech Lead
I've contributed to 50+ mobile games and prototypes, all published to the app stores and tested at least once. Here are a few of the ones I enjoyed working on the most. Click any project for the breakdown.
We needed to run boolean mesh operations in real time. With the Clipper2 library (by Angus Johnson) we did the boolean math in 2D, then extruded the result into a 3D mesh. The real challenge I solved was smoothing the carved tunnel and tuning the input feel, so the cut stays satisfying under a finger drag while keeping performance mobile-friendly.
A custom Unity Editor tool that lets artists select a group of gems and apply a gradient across them. It's driven via MaterialPropertyBlock, so it stays batching-friendly and removes an artist bottleneck the performance-correct way.
Themes were tied to player progression, not to levels — so I decoupled them and built tooling to apply any theme to any level. Kept content flexible, since different players could unlock a new theme at different points depending on their progression.
Outcome The standout among the many original prototypes we built and tested with CrazyLabs — it broke past the early CPI gates that kill most prototypes and scaled into the thousands before hitting limits short of 1M. It was also where I first owned monetization to this depth, building the full IAP and ad integration.
I built the jump-trick system to pick tricks from launch direction + approach angle + joystick input, so the same ramp produces varied, readable stunts. The hard part was a strict rule: the car must never land on its roof or bug out mid-jump. So I separated the physics movement system from the trick visuals to guarantee a clean and playable touchdown.
I drove zombie movement and crowding with Unity NavMesh, keeping large groups moving believably toward the player without tanking performance. The map was endless, built from random chunks that enable and disable dynamically, so I baked a NavMesh per chunk and stitched them together at runtime as chunks streamed in and out.
I made a custom Editor tool for modular ramp sizes so designers could author and resize ramps quickly, without needing new meshes modeled in Blender. It let level designers build new chunks without constantly depending on artists.
Outcome A spectacle-driven hybrid title where the core thrill — a perfect airborne smash — is guaranteed to feel good every time by design, not by luck.
Every block the player drops mutates the navigable grid, so I built pathfinding that recalculates on each placement, fast enough to feel instant on mobile. And since a TD breaks if players can wall off the base, I made a fully-sealed path impossible, checked against every placement.
I used projectors for splatter when enemies are crushed, and a LineRenderer to predict the enemy route from spawn to base, so intent is readable before you commit a block.
I helped build the meta-progression screen that lives outside the core loop, where players upgrade individual towers and choose which towers go into their deck. It turns single runs into a longer arc: a reason to come back and somewhere to spend what you earn.
Outcome One of our first hybrid-casual titles. The systems work (mutable navigation + guaranteed-solvable paths) is the part I'm proudest of.
The feel of cleaning is the product, so I built it in-shader. The first tool was a power-wash that uses a maskable dirt layer, removed exactly where the water touches the mesh. I then added a brush for scraping, a vacuum, and a sponge that forms bubbles and smears the dirt layer. Tactile, real-time, no fake animations.
I architected the cleaning systems to be reusable, so new objects could be set up quickly. That let us test objects trending in TikTok ASMR cleaning videos at the time, like the pool cleaning shown here, and even swap the inanimate object for a live animal to stand up a sibling game, Pet Salon DIY, on the same foundation.
Outcome I built both games almost entirely solo. Deep Clean DIY hit a strong CPI, and Pet Salon DIY pushed it even further, proof I can own a product end-to-end and hit the metric that decides whether it ships. It stalled for market reasons rather than product ones: the industry was shifting away from hyper-casual just as it found traction.
20M+ downloads
Paint-scatter feature used in bonus levels + in-Unity creative recording (joined post-launch, helped scale).
Roguelite
Life-bubble shader that transforms a barren world into a lush one as influence expands.
Classic TD
Full tower-defense core loop: unit placement, merging & upgrading, shooting, and enemy stacking.
Shipped with
I'm a game developer based in Slovenia who cares most about one thing: how a game feels to play, and building the systems that make that feel possible.
I've been making things with Unity and Blender since I was a kid, self-taught, long before it was a job. I went to a computer-science high school, then earned my BSc in Computer Science at the University of Ljubljana (FRI). Formal study was the closest path to the game development I wanted to do, and I never stopped building on the side.
In 2020 I joined Dawn of Play (the game studio at Razum d.o.o.) as a junior Unity developer. Since then I've grown into a lead role and helped the team scale from 3 to 8 people. I currently work as an independent contractor through my own company.
Along the way I've shipped across the mobile spectrum: hyper-casual with CrazyLabs (including I Can Paint, which the team scaled past 20M downloads), then hybrid-casual with Voodoo, Homa, and Hypercell, with deeper meta and more advanced core loops, trending toward midcore games.
What I'm actually good at: at the core I'm a Unity developer, with most of my experience there. On top of that, I know enough 3D, animation, and shaders to act as a bridge between the art and engineering sides, and I love building custom editor tools so artists, level designers, and game designers can do their best work without waiting on a programmer. I research new workflows and teach them through workshops and tutorials, and I encourage clear specs and documentation.
I also think about the whole pipeline, not just the code: UA and marketing, CPI testing, retention and playtime, feature A/B testing, shipping, live updates, monetization, and analytics. I believe in a build-measure-learn loop, and I like turning raw data into something a team can actually act on.
Outside work, you'll usually find me running, hiking, or on a padel court.
Unity (senior, shipped) · C# · Python · SQL
Shader Graph · Particle systems · Haptic feedback · Game-feel & Input tuning
Blender (modeling, UV, rigging) · Substance Painter · Photoshop · Figma · Bridging art & dev
Merging · Deck-building · Placement · Combat · Inventory · Upgrades · Painting · Cleaning · Cutting
Gameplay architecture · Data-driven & decoupled design · Reusable systems · Runtime navigation · Procedural generation · Multiplayer · Save systems · Editor tooling
Linux servers · Caddy/Nginx reverse proxy · Website & app deploys · Self-hosting (game servers, blockchain nodes, SpacetimeDB, databases)
AdMob · Firebase · IAP · GameAnalytics · Facebook SDK · Smartlook · GDPR · CPI & A/B testing · Funnels/KPIs
Always up for talking games, tools, and craft. Based in Slovenia.