Noppanat Wadlom
Portrait of Noppanat Wadlom

Noppanat Wadlom

Ph.D. Candidate in Computer Science at the National University of Singapore

Machine learning systems, big data systems, agentic AI

EmailScholarGitHubLinkedIn

About

I am a Ph.D. Candidate in Computer Science at the National University of Singapore (NUS), supported by the President's Graduate Fellowship. My research interests include machine learning systems, big data systems, and agentic AI, with a focus on how to execute complex multi-step and multi-agent LLM workflows efficiently on modern hardware.

Before NUS, I completed a B.Sc. in Computer Science (double major in Electrical Engineering) at KAIST with a 4.20/4.30 GPA. My applied work also includes causal modeling for banking analytics and production ML pipelines.

Education

National University of Singapore (NUS), Ph.D. in Computer Science

Aug 2024 – Present, Singapore

President's Graduate Fellowship (full support for exceptional research promise). Research focus: high-throughput LLM serving, workflow-aware scheduling, cache reuse across agentic steps.

Korea Advanced Institute of Science and Technology (KAIST), B.Sc. in Computer Science

Aug 2020 – Aug 2024, Daejeon, South Korea

Double major in Electrical Engineering. Semiminor in Artificial Intelligence.
4.20
/ 4.30 GPA
#1
of 90 in dept.
#3
of 258 in class
5
semesters on Dean's List

Experience 6

Computer Architecture and Systems Laboratory, KAIST Undergraduate Student Intern

Feb 2024 – Jun 2024, Daejeon, South Korea

Worked on efficient LLM serving, including speculative decoding and throughput/latency trade-offs in vLLM-style inference setups. Designed an algorithm to choose draft token counts for speculative decoding to improve serving efficiency.

Cyber-Physical Systems Laboratory, KAIST Undergraduate Research Participation

Jun 2023 – Feb 2024, Daejeon, South Korea

Built an AI-based UI Detection and Distribution system:
  • Trained and deployed real-time UI element detectors.
  • Built the extraction pipeline (vision inference, a lightweight Python server using multiprocessing and sockets, downstream delivery).
  • Optimized runtime behavior.

Siam Commercial Bank Data Science Intern, Data Analytics Function

Jun 2023 – Aug 2023, Bangkok, Thailand

Built a multi-task neural network model for conversion rate estimation with a causal debiasing objective. Designed data prep and feature pipelines for large-scale model training using Apache Spark and Azure Databricks.

Cyber-Physical Systems Laboratory, KAIST Undergraduate Student Intern

Dec 2022 – Jun 2023, Daejeon, South Korea

Extended a UI offloading and distribution system inspired by FLUID-XP.
  • Studied an Android OS variant for UI streaming.
  • Built a cross-platform client in Flutter.
  • Implemented peer-to-peer connectivity (Wi-Fi Direct, Bluetooth discovery).

Rayong Engineering and Plant Service Co., Ltd. Student Intern

Jun 2022 – Aug 2022, Rayong, Thailand

Developed a machine vision "Pellet Analyzer" for plastic manufacturing quality control:
  • Collected and labeled production images.
  • Trained an object detection model.
  • Shipped an interface for operators on the line.

Concurrency and Parallelism Laboratory, KAIST Individual Study Student

Dec 2021 – Jun 2022, Daejeon, South Korea

Studied persistent memory models, concurrent data structures, and file systems. Implemented a log-structured file system for rv6 (an xv6-style OS in Rust) to explore concurrency and durability behavior.

Publications 6

Efficient LLM Serving for Agentic Workflows: A Data Systems Perspective

SIGMOD 2026
N Wadlom, J Shen, Y Lu. Proc. ACM Manag. Data 4, 3, Article 169 (June 2026). arXiv:2603.16104DOI

mCardiacDx: Radar-Driven Contactless Monitoring and Diagnosis of Arrhythmia

IEEE JTEHM 2026
A Kumar, N Wadlom, J Kwak, SH Kang, I Shin. arXiv:2508.02274DOI

Lumilake: An Agentic Analytics Engine for AI4Science

Preprint 2026
Z Su, N Wadlom, J Shen, P Zhang, R Luo, D Liu, Y Hou, Y Huang, W Wu, Y Lu. PDF

FlowMesh: A Service Fabric for Composable LLM Workflows

arXiv 2025
J Shen, N Wadlom, L Zhou, D Wang, X Miao, L Fang, Y Lu. arXiv:2510.26913

Batch Query Processing and Optimization for Agentic Workflows

arXiv 2025
J Shen, N Wadlom, Y Lu. arXiv:2509.02121

Glioma Brain Tumor Classification using Transfer Learning

ICEAST 2024
P Pilaoon, A Narkthewan, N Wadlom, R Varakulsiripunth, K Hamamoto, et al.

Awards and honors 4

President's Graduate Fellowship, National University of Singapore

2024

Competitive fellowship awarded to top Ph.D. students showing exceptional research potential. Covers full tuition and stipend at NUS.

College of Engineering Dean's List, KAIST

2021 – 2024

Awarded to top-performing students (about the top 3%) nominated by the department. Fall 2021, Fall 2022, Spring 2023, Fall 2023, Spring 2024.

KAIplus+ Scholarship, KAIplus

2022

Merit scholarship awarded to a small number of international KAIST students each semester.

Honorable Mention, International Mathematical Modeling Challenge

2019

National team recognition for high-quality applied mathematical modeling work.

Teaching 2

CS4262/5462 Machine Learning Systems Teaching Assistant

AY2025/26 Sem 2, NUS

Undergraduate and graduate course. Instructor: Yao Lu.

CS6216 Advanced Topics in Machine Learning Teaching Assistant

AY2025/26 Sem 1, NUS

Graduate course. Instructor: Yao Lu.

Skills

Programming
Python, C, Java, Dart; also C++, Rust, Scala, JavaScript
ML, data, systems
PyTorch, TensorFlow, scikit-learn, Pandas, NumPy, Apache Spark, Spark ML, Azure Databricks, MySQL, Flask, Flutter
Research areas
LLM serving efficiency, agentic workflows, speculative decoding, KV cache optimization, query optimization
Languages
Thai (native), English (fluent), Korean (beginner)

Contact

Email
noppanat@comp.nus.edu.sgnoppanat@u.nus.edu
Phone
(+65) 8098 6325
Based in
Singapore and Thailand
EmailScholarGitHubLinkedIn