Kaan Sancak (original) (raw)
Senior Research Scientist at Meta 
I work on machine learning systems, recommendation, and AI infrastructure. My background is in high-performance computing and machine learning, and I am interested in the systems and algorithms that make modern ML efficient, scalable, and useful in practice.
Before Meta, I received my Ph.D. in Computer Science fromGeorgia Tech 
, advised byUmit V. Catalyurek, where I focused on high-performance computing, GPU systems, parallel algorithms, and machine learning at scale.
Email Google Scholar GitHub LinkedIn CV
🔬 Research Interests
High-Performance Computing Machine Learning ML Systems & AI Infrastructure Recommendation Systems Parallel & Distributed Computing GPU Computing
💼 Experience
Meta Full-time · Aug 2024 – present
Senior Research Scientist Meta Recommendation Systems (MRS) Feb 2026 – present
Building next-generation recommendation models and ML infrastructure powering Meta's core ranking and personalization systems at scale.
Research Scientist Ads Ranking & Foundational AI (RAI) Aug 2024 – Feb 2026
Model–infrastructure co-design for billion-scale ad recommendation. Built real-time graph integration improving data freshness from days to minutes, boosted training throughput by 20%, and cut feature storage cost by 3x. Lead contributor to the Ads Graph Foundational Model (GFM).
Meta Internship · 2023
Research Scientist Intern Ranking & Foundational AI May – Aug 2023
Conducted research on scalable graph-based models for efficient learning without sacrificing quality. Work published at AAAI 2025 and ICLR 2024.
Meta Internship & Part-time · 2022
Part-time Student Researcher AI Systems HW/SW Co-Design Aug – Dec 2022
Extended caching mechanisms for Meta's ML training platform, substantially reducing redundant data serving computations for key models. See the AI System Co-design project.
Research Scientist Intern AI Systems HW/SW Co-Design May – Aug 2022
Built caching infrastructure for Meta's data ingestion pipelines, eliminating redundant computation across large-scale model training runs. See the AI System Co-design project.
Pacific Northwest National Laboratory Internship · 2021
Research Intern HPC & Systems May – Aug 2021
Distributed graph algorithms and high-performance data structures on the SHAD framework.
Facebook Internship · 2020
Research Intern AI Systems HW/SW Co-Design May – Aug 2020
Improved Facebook's graph engine performance via novel partitioning — 10% query throughput gain, up to 5x end-to-end speedup. Integrated the engine with Instagram Ads; infrastructure still serves Instagram, Threads, and Facebook.
Google Summer of Code / NRNB 2018
Open Source Developer May – Aug 2018
Built collaborative pathway editing tools forcBioPortal for Cancer Genomics (Memorial Sloan Kettering Cancer Center).
IBM Internship · 2017
Software Engineering Intern Jun – Aug 2017
Cloud data transfer and object recognition apps (IBM Cloud, Python, Kafka).
🎓 Education
Georgia Institute of Technology
GPA: 4.00/4.00 · Advisor: Umit V. Catalyurek
Research High-performance computing, machine learning, graph systems and algorithms
Teaching CSE 6230: High-Performance Parallel Computing · CSE 6220: Intro to HPC
Bilkent University, Turkey
Summa Cum Laude · GPA: 3.82/4.00 · Ranked 4th / 231 engineering students
📖 Selected Publications
Haystack Engineering: Context Engineering for Heterogeneous and Agentic Long-Context Evaluation
M. Li, D. Fu, L. Wang, S. Zhang, H. Zeng, K. Sancak, R. Qiu, H. P. Wang, X. He, X. Bresson, Y. Xia, C. Sun, P. Li
arXiv 2025 Paper
Haystack Engineering: Context Engineering Meets the Long-Context Challenge in LLMs
M. Li, D. Fu, L. Wang, S. Zhang, H. Zeng, K. Sancak, R. Qiu, H. P. Wang, X. He, X. Bresson, Y. Xia, C. Sun, P. Li
NeurIPS 2025 Workshop Paper