Ilker Kesen
Postdoctoral Scientist at Department of Computer Science, University of Copenhagen - DIKU
Copenhagen, Capital Region of Denmark
Summary
🤩
Rockstar🎓
Top SchoolIlker Kesen is a postdoctoral scientist at the University of Copenhagen with 15 years of experience spanning research, software engineering, and teaching in machine learning and language-vision integration. He earned a PhD from Koç University and has a strong practical track record building and reproducing deep learning models—contributing examples and bug fixes to the Knet.jl framework to improve VGG and ResNet CPU compatibility and batch normalization correctness. His work bridges academia and applied engineering through dataset development for video-language models, metric-driven optimization projects, and hands-on Julia implementations for research reproducibility. Comfortable mentoring and lecturing, he has taught graduate deep learning and NLP courses while tutoring teams in Julia at research institutes. Based in Copenhagen, he combines rigorous research instincts with pragmatic software craftsmanship, often surfacing subtle implementation issues that make models robust across environments.
15 years of coding experience
3 years of employment as a software developer
Doctor of Philosophy (PhD) Computer Science and Engineering, Doctor of Philosophy (PhD) Computer Science and Engineering at Koç UniversityBachelor of Science (B.S.) Electronics and Communication Engineering, Bachelor of Science (B.S.) Electronics and Communication Engineering at Istanbul Technical University
Stackoverflow
483reputation
753kreached
8answers
2questions
Github Skills (17)
machine-learning10
deep-learning10
neural-network10
knexjs10
knn10
julia10
vggnet9
resnet9
data-science8
rfc6
iso6
rest6
tornado6
websocket6
python6
Programming languages (7)
Github contributions (5)
denizyuret/Knet.jl
Nov 2016 - Dec 2018
Koç University deep learning framework.
Role in this project:
ML Engineer
ML EngineerContributions:55 commits, 23 PRs, 1 push in 2 years 1 month
Contributions summary:Ilker focused on enhancing the VGG and ResNet example implementations within the Knet deep learning framework. Their contributions included making the VGG inference example work for both D and E models and adding a ResNet example. They also addressed issues related to CPU compatibility and corrected errors in the batch normalization layer, indicating an effort to ensure the models function correctly across different environments. Further refinements involved fixing a model selection error and updating examples to use the latest version of Images.jl.
data-sciencedeep-learningknetneural-networksmachine-learning
ITURO/ituro
Mar 2015 - Apr 2017
ITURO Core Systems.
Contributions:109 commits, 6 PRs, 118 pushes in 2 years 1 month
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.