Ilmari Heikkinen
Managing Director at Heichen Limited
Hong Kong, China
Summary
🤩
Rockstar🎓
Top SchoolIlmari Heikkinen is a Managing Director and creative technologist with 18 years of experience building immersive “phygital” art and retail installations that fuse WebGL, Unity, AR, sensors, cameras and AI into large-scale interactive experiences. As founder of Heichen studio he’s delivered award-winning projects for Chanel, LVMH, Google and the M+ Museum, and was a Top Finalist for the LVMH Innovation Award while contributing to SXSW 2025 winner Synegram.com. An early WebGL spec contributor at Khronos and a former Google Developer Programs Engineer, he uniquely blends standards- and developer-facing credentials with hands-on art direction. His open-source work spans practical WebGL/AR tooling and cutting-edge ML optimizations (notably memory/performance enhancements to HuggingFace Diffusers enabling very high-resolution generation). He also engineers deep-performance systems—RDMA file transfers, GPU file IO and compute-shader libraries—which underline a rare mix of creative vision and low-level systems skill. Based in Hong Kong, he builds novel production pipelines that turn complex technical constraints into accessible, mesmerizing experiences.
18 years of coding experience
8 years of employment as a software developer
Bachelor of Science (BSc), Computer, Bachelor of Science (BSc), Computer at University of HelsinkiJoutsenon lukio
Finnish, English, Japanese, Swedish, German
Github Skills (18)
pytorch10
javascript10
diffusion-models10
stable-diffusion10
diffusers10
diffusion-probabilistic-models10
diffusion-probabilistic10
deeplearning-ai10
deep-learning10
image-generation10
python9
html9
cuda8
webcam8
mediastreamtrack8
Programming languages (8)
Github contributions (5)
kig/JSARToolKit
Mar 2011 - Mar 2016
JavaScript port of FLARToolKit
Role in this project:
Front-end Developer
Front-end DeveloperContributions:15 commits, 4 PRs, 4 pushes in 5 years 1 month
Contributions summary:Ilmari primarily focused on updating and modifying the JavaScript code, likely related to the core functionality of the project. They updated bundled files, fixed bugs, and implemented changes to the demo files. Their contributions include code modifications in `JSARToolKit.js` and enhancements to the example slideshow for webcam use.
jsjavascriptjsinterop
huggingface/diffusers
Nov 2022 - Dec 2022
🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch and FLAX.
Role in this project:ML Engineer
ML EngineerContributions:3 reviews, 3 commits, 5 PRs in 21 days
Contributions summary:Ilmari's contributions primarily focus on enhancing the `diffusers` library, specifically addressing performance and memory optimizations within the Stable Diffusion pipeline. Their work includes implementing VAE slicing for larger batch sizes, integrating xFormers for efficient attention mechanisms, and introducing tiled VAE decoding for high-resolution image generation. These changes aim to improve memory usage and overall performance in the image generation process.
pytorchartdeep-learningimage2imagestate-of-the-art
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.