Samet Akcay
Principal AI Research Engineer at Intel Corporation
England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolSamet Akcay is a Principal AI Research Engineer at Intel with nine years of experience building real-time image classification, detection, and self-supervised anomaly detection systems for industrial, medical, and security applications. He led core contributions to anomalib, one of the largest open-source anomaly detection libraries, and has hands-on experience refactoring data pipelines and edge inference tooling for production use. His background spans academia and industry—PhD in Computer Science from Durham and an MSc from Penn State—where his work funded by UK government agencies translated into practical CV solutions. Previously he architected self-supervised anomaly modules for COSMONiO’s NOUS platform, the world’s first interactive deep learning box for subject-matter experts. Known for bridging rigorous research with production engineering, he often focuses on unsupervised feature learning and dataset engineering that make anomaly models robust and deployable.
9 years of coding experience
12 years of employment as a software developer
Master of Science (MSc) Electrical Engineering, Master of Science (MSc) Electrical Engineering at Penn State UniversityDoctor of Philosophy (PhD) Computer Science, Doctor of Philosophy (PhD) Computer Science at Durham UniversityBachelor's Degree Electrical and Electronics Engineering, Bachelor's Degree Electrical and Electronics Engineering at Gazi University
English, Turkish
Github Skills (19)
pytorch10
preprocessing10
anomaly-detection10
cifar10010
python10
dcgan10
preprocess10
machine-learning10
cgan10
semi-supervised-learning10
data-preprocessing10
data-loading10
dataprep10
mnist10
data-processing10
Programming languages (3)
Github contributions (5)
samet-akcay/ganomaly
May 2018 - Feb 2022
GANomaly: Semi-Supervised Anomaly Detection via Adversarial Training
Role in this project:
ML Engineer
ML EngineerContributions:68 commits, 11 PRs, 47 pushes in 3 years 9 months
Contributions summary:Samet primarily contributed to the data loading and model training aspects of the GANomaly project. They modified the CIFAR10 dataloader to incorporate anomaly detection, enabling the training process to focus on specific anomaly classes. Furthermore, the user added MNIST dataset support, and implemented a new dataset structure which included the get_mnist_anomaly_dataset and get_mnist2_anomaly_dataset methods. They also incorporated features that support the use of random seeds.
pytorchanomalysuperviseddeep-learningadversarial
open-edge-platform/anomalib
Nov 2021 - Jan 2023
An anomaly detection library comprising state-of-the-art algorithms and features such as experiment management, hyper-parameter optimization, and edge inference.
Role in this project:Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:33 releases, 1298 reviews, 215 commits in 1 year 2 months
Contributions summary:Samet primarily refactored and refactored the code related to the `anomalib.dataset` module, moving the dataset-related code to `anomalib.data` to better organize the project structure. They also made code changes to the MVTec dataset, including adding and refactoring the dataset. The user appears to be working on data-related tasks and contributing to the data loading and processing pipeline.
hyper-parameter-optimizationstate-of-the-arthyperdetection-libraryexperiment-management
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.