About
Hello! I’m Daniel, a Master’s student in Computer Science at the University of Pennsylvania.
I am currently working with Prof. Vincent Liu in the Distributed Systems Lab on techniques to accelerate LLM serving systems through dynamic scheduling strategies. My research focuses on systems for emerging workloads, particularly at the intersection of large-scale machine learning and systems design. Over the years, I have developed strong experience in both research and engineering for complex systems problems. Prior to Penn, I worked as a software engineer on the ShopeePay core team at Shopee. I received my bachelor’s degree from Harapan Bangsa Institute of Technology in Indonesia.
I’m actively seeking PhD opportunities in computer systems for Fall 2026. My research interests span systems-for-ML, ML-for-systems, datacenter computing, operating systems, storage systems, distributed systems, and databases.
Education
University of Pennsylvania
Master of Science in Engineering
Computer Science, 2024-2026
Advised by Prof. Vincent Liu in the Distributed Systems Lab
University of Pennsylvania
Visiting Student, 2021
Google Bangkit Academy
Machine Learning Bootcamp, 2020
Harapan Bangsa Institute of Technology
Bachelor of Technology
Informatics, 2018-2021
Preprints
Dynamic Rebatching for Efficient Early-Exit Inference with DREX
Xuting Liu, Daniel Alexander, Siva Kesava Reddy Kakarla, Behnaz Arzani, Vincent Liu
arXiv:2512.15705, 2025
Publications
Application of Convolutional Neural Network for Semantic Segmentation of Bandung Urban Scenes
Daniel Alexander, Hans Christian Kurniawan, Irfin Afifudin, Hery Heryanto
IEEE International Conference on Data and Software Engineering (ICoDSE), 2022
Selected Projects
DREX — Dynamic Rebatching for Early-Exit LLM Inference
Re-architects early-exit LLM serving by rebatching across exit depths to reduce KV-cache waste and tail latency.
PennOS — UNIX-like Operating System from Scratch
UNIX-like OS kernel implemented in C, featuring preemptive scheduling, virtual memory, IPC, and a FAT-based file system.
PennCloud — Fault-Tolerant Distributed Cloud Platform
Distributed email and file-storage platform built on a custom replicated key-value store with strong consistency and transparent failover.
Mini Minecraft
Voxel-based engine built in C++/OpenGL with custom shaders, physics, collision handling, and GUI systems.
Teaching Experience
CIS4190/5190 Applied Machine Learning
Head Teaching Assistant, 2025
Teaching Assistant, 2024
Industry Experience
Shopee
Software Engineer, ShopeePay Core Team
2022-2024
Honors & Awards
Master's Scholarship - Full Funding
University of Pennsylvania Master's Program
Indonesia Ministry of Education and Research
<2% acceptance rate
2024
Student Mobility Scholarship - Full Funding
University of Pennsylvania visiting student
Indonesia Ministry of Education and Research
<1% acceptance rate
2021
Google Bangkit Academy - Best Capstone Project
National Winner
<8% admission rate, <2% selection rate
2020