Software Engineer · AI
Exploring the intersection of engineering craft, artificial intelligence, and the geometry of understanding.
Selected Work
Calculating overspray in industrial ceramic coating using Alpha Shape analysis and Möller-Trumbore ray-triangle intersection on 3D point cloud data. Developed as part of the 3D-ProSim project at GFaI Berlin. Presented at the 3D-NordOst workshop.
Read the story
Binary classification of mouse well-being from facial images using ResNet18 and ConvNeXt-Tiny. Compared muzzle crops versus full-face images across 14 experimental configurations with Grad-CAM and saliency map analysis.
Read the story
Bachelor's thesis on semantic segmentation of 3D LiDAR forest point clouds using gradient-boosted tree classifiers. Interpreted geometric feature effectiveness with SHAP explainability — global vs local attribution on XGBoost, LightGBM, and CatBoost models.
Read the story
Research into machine-learning approaches for predicting temperature and porosity evolution in cast metal components, with the goal of accelerating computationally expensive casting simulations.
Read the story
A mobile application for discovering nightlife venues, employing Supabase for data storage and authentication. Developed independently and deployed to the App Store.
An internal automation tool built at GFaI to automate birthday messages for team members. It improved a repetitive workflow by delivering consistent, personalized celebration messages to the team.
Read the story
Publications
About
My background is rooted in Geomatics Engineering, where I first became interested in how geometry, spatial data, and computation can be used to understand the physical world.
After studying Geomatics Engineering in Istanbul, I continued with a Master's in Geodesy and Geoinformation Science at TU Berlin, focusing on computer vision, remote sensing, image processing, geographical information systems and spatial data analysis. Along the way, my interests gradually moved from understanding physical environments to teaching machines how to interpret them.
Today, I work at the intersection of computational geometry, computer vision, and artificial intelligence. At GFaI, I develop machine-learning and computer-vision systems for industrial applications, working with 3D point clouds, digital images, spatial data and robotics.
I enjoy working on problems where software meets the physical world — turning complex spatial data into systems that can perceive, reason, and solve practical problems.
Currently open to research collaborations, interesting problems, and thoughtful conversations.
hello@burakhafizoglu.com