Dogukan portrait

Hello, I'm Dogukan Kaan Bozkurt. I'm a Senior Software Engineer, Ads & Growth/ Playable Ads Developer/ HTML5 Game Developer. I enjoy building interactive ads, crafting AI-powered automation tools, and turning campaign data into growth.

About me

I'm a Software Engineer working at the intersection of ads, games, and growth, with 5 years of experience building playable ads, HTML5 games, and the tooling behind them. My work spans the full lifecycle of an ad unit; from mechanic design and 2D/ 3D development to network packaging and post-launch analysis of ROAS, IPM, CTR, CVR, and retention.

I've shipped playables for mobile games with tens of millions of installs; my most recent playable set ran 320+ variants across 7 languages and scaled to $560K+ in daily ROAS-campaign spend, with 14M+ daily impressions and 1M+ daily clicks. I work with TypeScript, JavaScript, C#, Unity, Luna Playworks, PixiJS, and Three.js. This background has reinforced my appreciation for clean architecture, OOP principles, and maintainable codebases.

Beyond the ads themselves, I build AI-powered tools, Node.js platforms, and npm libraries that automate creative production, and I use campaign and experiment data to drive technical decisions. I'm continuing to grow toward ad platforms and AdTech engineering: experimentation, measurement, and large-scale systems that make advertising work better for users and advertisers.

Experience

logo

Agave Games

Istanbul, Turkey

Apr. 2026 - Present

Senior Software Engineer, Performance Marketing

Apr. 2026 - Present

  • Owned end-to-end playable ad development for a 50M+ install puzzle portfolio, scaling one title to 500+ variants, driving $560K+ in daily ad spend from playables alone and 1M+ daily clicks.
  • Collaborated with PMs, growth managers, and artists on rapid gameplay iteration, A/B testing, and competitor benchmarking to improve CTR, IPM, and CVR while preserving gameplay accuracy and early retention.
  • Co-built Agave AI Studio, an internal AI creative platform replacing a third-party UGC tool at ~1/3 the cost, featuring AI video generation, localization, and an LLM agent converting gameplay videos into playables.
  • Built a JSON-driven web templating system enabling designers and growth managers to create and ship playable variants independently, accelerating high-volume creative testing.
logo

JustDice

Hamburg, Germany

Apr. 2023 - Apr. 2026 2 positions

  1. Senior Playable Ads Developer

    Oct. 2024 - Apr. 2026

    • Owned playable ads and IECs end-to-end across a rewarded-gaming portfolio of 20+ apps and 300M+ lifetime downloads, including the flagship title Cash Giraffe (10M+ installs), from competitor benchmarking and concept development to post-launch optimization of CTR, IPM, CVR, and ROAS.
    • Built an internal TypeScript/Node.js platform for producing playable ads, HTML5 games, and IECs, featuring automated asset compression and one-click multi-network export. Also developed AI-powered plugins for Photoshop, After Effects, and Figma, which were rolled out across teams, reducing production time per creative.
    • Analyzed A/B test results and in-playable interaction-event data to identify engagement patterns and drop-off points throughout the ad-to-install funnel, aligning ad messaging with the in-app experience to improve install rates and user quality (retention and LTV).
  2. Playable Ads Developer

    Apr. 2023 - Oct. 2024

    • Developed size-optimized playable ads and HTML5 games for rewarded-gaming apps, aligned with UA strategy and optimized for CTR, IPM, and CVR.
    • Created performance-optimized 2D/3D assets using Blender, Photoshop, and Figma, balancing visual quality with strict ad-network size limits.
    • Built plugins for Unity, Photoshop, After Effects, and Figma to automate marketing production workflows and accelerate creative production.
logo

MY.GAMES

Amsterdam, The Netherlands

Mar. 2025 - Mar. 2026

Playable Ads Developer & Consultant (External)

Mar. 2025 - Mar. 2026

  • Designed and developed playable ads for mobile games in TypeScript, Three.js, and PixiJS, MRAID-compliant creatives within each ad network's size and CTA rules for global UA campaigns.
  • Advised on the technical and creative approach for each playable, optimizing load time and runtime performance on low-end devices and tightening core gameplay hooks to lift CTR and IPM.
logo

GameGame

Hamburg, Germany

Jul. 2025 - Nov. 2025

Playable Ads Development Consultant

Jul. 2025 - Nov. 2025

  • Established the company's first playable ad pipeline from the ground up, defining the tech stack, build tooling, and QA/delivery workflow that took the team from zero to shipping network-ready playables for UA campaigns.
logo

Panteon Games

Ankara, Turkey

Nov. 2021 - Apr. 2023

Playable Ads & Marketing Game Developer

Nov. 2021 - Apr. 2023

  • Developed size-optimized Unity/Luna playable ads and HTML5 games for hyper- and hybrid-casual titles, supporting global UA campaigns.
  • Rapidly prototyped gameplay mechanics in Unity (C#) for video creatives and marketability tests, validating hooks through CTR and IPM performance.
  • Built reusable playable ad templates, Unity packages, and editor tools to streamline production and accelerate creative A/B testing across ad networks.
  • Collaborated with UA, growth, and creative teams to optimize ad performance (IPM, CPI, CTR) while mentoring junior developers on production workflows.
logo

TUPRAS

Kocaeli, Turkey

Aug. 2021 - Sep. 2021

Project Engineer - Intern

Aug. 2021 - Sep. 2021

  • Led 'susTRAINable' project at TUPRAS, targeting net-zero railway emissions by 2050, aligning with 6 UN Sustainable Development Goals.
  • Led 'TogetHER' project focused on improving women's lives in the refinery and energy sectors, including mentorship programs for STEM-inclined women.
logo

GEFEASOFT

Mugla, Turkey

Mar. 2021 - Jul. 2021

Game Developer - Intern

Mar. 2021 - Jul. 2021

  • Developed WebGL and HTML5-based serious games using C# and Unity.
  • Designed gameplay systems, optimized performance for web deployment, and participated in feature development throughout the project lifecycle.
logo

Avocuda

Istanbul, Turkey

Feb. 2020 - Jun. 2020

Software Engineer - Intern

Feb. 2020 - Jun. 2020

  • Developed and maintained frontend features for mobile applications as part of the engineering team.

Education

Skills

- Technical Skills -

  • C# icon
    C#
  • TypeScript icon
    TypeScript
  • JavaScript icon
    JavaScript
  • Python icon
    Python
  • C++ icon
    C++
  • HTML icon
    HTML
  • CSS icon
    CSS
  • TailwindCSS icon
    TailwindCSS
  • Sass icon
    Sass
  • React icon
    React
  • .Net icon
    .Net
  • Node.js icon
    Node.js
  • Express.js icon
    Express.js
  • MySQL icon
    MySQL
  • PostgreSQL icon
    PostgreSQL
  • MongoDB icon
    MongoDB
  • Jest icon
    Jest
  • Docker icon
    Docker
  • RabbitMQ icon
    RabbitMQ
  • Git icon
    Git
  • npm icon
    npm
  • Webpack icon
    Webpack
  • Vite icon
    Vite
  • UXP icon
    UXP
  • Unity icon
    Unity
  • Luna (Unity Playworks) icon
    Luna (Unity Playworks)
  • Cocos Creator icon
    Cocos Creator
  • ThreeJS icon
    ThreeJS
  • PixiJS icon
    PixiJS
  • GSAP icon
    GSAP
  • GLSL icon
    GLSL
  • React Three Fiber icon
    React Three Fiber
  • Blender icon
    Blender
  • Spline icon
    Spline
  • Figma icon
    Figma
  • Adobe PhotoShop icon
    Adobe PhotoShop
  • Adobe After Effects icon
    Adobe After Effects
  • Adobe Premier Pro icon
    Adobe Premier Pro

- Soft Skills -

Languages

Language Image

Turkish

Native

Language Image

English

C1

Language Image

Korean

A2

Language Image

German

A2

Playable Ads

Projects

Agave AI Studio

Aug. 2026 - Present

Co-built Agave's internal generative AI platform for image, video, lipsync, and localization ads at 1/3 the cost of a third-party UGC tool. Built an LLM agent (Claude, OpenAI) that turns gameplay videos into HTML5 playable ads.

  • Next.js
  • React
  • TypeScript
  • Node.js
  • MongoDB
  • GCP
  • LLM

smart-sheet-translate

Apr. 2026

Built a Google Sheets add-on in Google Apps Script that batch-translates game localization sheets through a custom Node.js API and the OpenAI API, delivering full multi-language coverage directly inside the spreadsheet.

  • Google Apps Script
  • Node.js
  • clasp
  • OpenAI API

base122-encoding (npm library)

Sep. 2025 - Nov. 2025

Built a TypeScript npm library implementing Base-122 binary-to-text encoding for data URIs, cutting encoding overhead from 33% (Base64) to about 14% and shrinking embedded assets in single-file HTML5 playable ads.

  • TypeScript
  • Node.js
  • Vite

texture-optimize-pro (npm library)

Oct. 2025

Built a TypeScript npm library that batch-optimizes textures for PixiJS and Three.js HTML5 games using Sharp, with per-texture max size, format, and quality settings to keep builds within ad-network size limits.

  • TypeScript
  • Node.js
  • Sharp

Figma Design Plugin

Jul. 2025 - Present

Built a Figma plugin in TypeScript that automates production of app store screenshots, ad-network end cards, and marketing assets, with DeepL API integration for one-click localization across multiple GEOs.

  • Figma
  • TypeScript
  • DeepL API
  • HTML
  • CSS
  • Vite

AdCraft

Oct. 2024

Built a React and TypeScript web tool that manages ad-network requirements and generates network-ready builds of playable ads and HTML5 games from a single interface, replacing manual per-network packaging.

  • React
  • TypeScript
  • Node.js
  • Tailwind CSS
  • Vite

Adobe Photoshop Design Tool

Jul. 2023 - Dec. 2023

Built an Adobe Photoshop plugin with UXP and ExtendScript that automates generation of app store screenshots, ad-network end cards, and marketing assets, reducing manual design work for the creative team.

  • UXP
  • ExtendScript
  • JavaScript
  • HTML
  • CSS

Unity Playable Ads Kit

Apr. 2022 - Sep. 2023

Built a Unity (C#) package for playable ad development with project template generation, a localization system, and reusable gameplay utilities, accelerating production of new playables and A/B test variants.

  • C#
  • Unity

Semantic Segmentation using Deep Learning

Sep. 2020 - Jun. 2021

Implemented deep learning models in PyTorch for 3D point cloud semantic segmentation, enabling a mobile robot to understand its surroundings in search-and-rescue scenarios at the university's AI and Robotics Lab.

  • Python
  • PyTorch
  • NumPy
  • pandas
  • SciPy
  • OpenGL

Interface Design for Segmentation Data

May 2021 - Jun. 2021

Built a Python desktop application with Qt and OpenGL to organize input data, train and test segmentation models, and visualize 3D point cloud results on the S3DIS and ESOGU RAMPS datasets.

  • Python
  • Qt
  • OpenGL

Point Cloud Data Optimization

Nov. 2020 - Jan. 2021

Developed a C++ point cloud downsampling algorithm that uses similarity measures and midpoint merging to reduce object data size while preserving the features needed for deep learning classification.

  • C++

Activities

Certificates

Three.js Journey

Jun. 2024

three.js journey by Bruno Simon

  • ThreeJS
  • JavaScript
  • React
  • React Three Fiber
  • HTML
  • CSS
  • Blender
  • GLSL
  • GSAP
  • CannonJS
  • Rapier

Back-End Apps with Node.js and Express

Dec. 2023

IBM

  • JavaScript
  • Node.js
  • Express

Professional Meta Front-End Developer Certificate

Nov. 2023

META

  • React
  • JavaScript
  • Bootstrap
  • HTML
  • CSS
  • npm
  • Jest
  • Figma
  • Sass

Intermediate Object-Oriented Programming for Unity

May 2021

University of Colorado

  • C#
  • Unity

Contact Me

Please contact me directly at dkaanbozkurt@gmail.com or through this form.