Resumé

Download PDF

RESEARCH STATEMENT

Advancing the generative AI through innovative approaches to generative models, with a focus on motion transfer in video generation, interpretability, and personalization techniques. My research aims to develop more controllable, interpretable, and efficient generative models that can bridge the gap between user intent and model output while maintaining high fidelity.

EDUCATION

AUG 2023 - MAY 2027 Virginia Tech, Blacksburg, VA Ph.D. in Computer Science

SEP 2018 - JAN 2022 Boğaziçi University, Istanbul, Turkey M.S. in Computer Engineering

SEP 2012 - JUN 2017 Boğaziçi University, Istanbul, Turkey B.S. in Computer Engineering

PUBLICATIONS

2025 ConceptAttention: Diffusion Transformers Learn Highly Interpretable Features ICML 2025 arXiv

2024 MotionFlow: Attention-Driven Motion Transfer in Video Diffusion Models arXiv preprint arXiv:2412.05275 Project Page

2024 MotionShop: Zero-Shot Motion Transfer in Video Diffusion Models with Mixture of Score Guidance arXiv preprint arXiv:2412.05355 Project Page

2024 CLORA: A Contrastive Approach to Compose Multiple LORA Models arXiv preprint arXiv:2403.19776 Project Page

2024 Conform: Contrast is all you need for high-fidelity text-to-image diffusion models CVPR 2024 Project Page

2024 Conditional Information Gain Trellis Pattern Recognition Letters, 184, 212-218 Publication Link

2022 Unsupervised Routing Strategies for Conditional Deep Neural Networks MSc Thesis. Boğaziçi University

2020 BURST: Software and Simulation Solutions of an Autonomous Vehicle 2020 28th Signal Processing and Communications Applications Conference (SIU) (pp. 1-4)

2018 Privacy-Preserving Intersection Management for Autonomous Vehicles Proceedings of the Tenth International Workshop on Agents in Traffic and Transportation (ATT 2018) (pp. 49-56)

RESEARCH EXPERIENCE

MAY 2025 - UPCOMING Amazon Applied Scientist Intern

• Will join Amazon’s AGI team working on auto-regressive generative model.

AUG 2023 - CURRENT Virginia Tech Research Assistant and Lab Lead

• Lead research efforts on generative models, focusing on enhancing diffusion-based text-to-image, text-to-video models and interpretability methods.
• Collaborate with Google to implement research findings in closed-source diffusion-based image generation models. The work has been published at CVPR 2024 and subsequent work submitted to SIGGRAPH 2025.
• Awarded Deloitte Research Fellowship to work on mechanistic interpretability of large language models by learning steering vectors in an unsupervised way.
• Research on video diffusion models to efficiently transfer motion from real videos to generated videos.

MAY 2024 - AUG 2024 Adobe Research Intern in Video Generative AI

• Developed and refined instruction-based video editing methods using advanced video diffusion transformers and multi-modal image-text datasets.
• Designed and implemented methods to leverage temporal capabilities of video models for enhancing video editing workflows.

PROFESSIONAL EXPERIENCE

Nov 2022 - AUG 2023 Lyrebird Studio Machine Learning Engineer

• Developed and maintained image generation ML services handling 5 million daily requests.
• Architected robust machine learning CI/CD pipelines using GitHub Actions and utilized AWS CDK for building architecture as code, enabling seamless deployment of research team outputs as production-ready services.
• Led the design and deployment of diffusion-based model training and image generation services, effectively handling thousands of daily requests on GPU-accelerated instances with high performance and stability.
• Integrated Generative AI solutions into applications, enhancing user experience.

AUG 2021 - Nov 2022 Vispera Machine Learning Engineer

• Spearheaded the automation of deep learning model training using Python and TypeScript, resulting in a tenfold increase in daily model deployments, significantly reducing development time and costs.
• Launched a user-friendly VueJS front-end service empowering researchers to train and deploy new models by providing real-time monitoring of online and offline metrics, enhancing model observability and productivity.
• Worked as a full-stack machine learning engineer, using VueJS in frontend services; Python in machine learning services; TypeScript, NodeJS, Go, PostgreSQL, and MongoDB in backend services.

Oct 2019 - AUG 2021 Vispera Computer Vision Research Engineer

• Led research and development for deep learning image recognition models, utilizing Python, TensorFlow, and OpenCV, to solve challenging problems related to out-of-distribution recognition and hierarchical classification.
• Pioneered the formulation and implementation of a novel zero-shot learning-based image recognition model using PyTorch, which significantly reduced image annotation time by four times. This innovative approach recommends best matches without annotated data, optimizing the model development process.

AUG 2017 - Oct 2019 Idea Technology Solutions Computer Vision Research Engineer

• Developed novel tree-based deep learning architectures, improving performance in object detection.

HONORS & AWARDS

2025 (IN PROGRESS) FINALIST, Qualcomm Innovation Fellowship

2020 RUNNER-UP, Kaggle Anadolu Sigorta Datathon Challenge

2019 & 2020 FINALIST, Teknofest Autonomous Vehicle Competition

2019 & 2020 WINNER, Teknofest Autonomous Vehicle Competition - Simulation

2018 WINNER, Mercedes-Benz Hackathon

2017 WINNER, BSH Analytics for Production Excellence Hackathon

WORKSHOPS & OUTREACH

Oct 2025 ICCV 2025 - Personalization in Generative Models Organizer

• Leading organization efforts for an international workshop focused on personalization techniques in generative AI.

JUN 2021 inzva - METU ImageLab AI Labs Joint Program Guide

• Conducted lectures of probability, statistics, and graphical models for the Deep Generative Models course, organized in collaboration with Prof. Gokberk Cinbis from METU.

JUL 2018 - DEC 2020 Boğaziçi University Autonomous Vehicle Team Founder

• Founded a team and laboratory for building an electric autonomous vehicle, creating autonomous driving R&D opportunities at Bogazici University.