hi , i am
andrew
wang .

CS student and Software Engineer

download resume

about me

I’m an MSCS student at Carnegie Mellon University. Before CMU, I studied Computer Science and Statistics at the University of Maryland, where I graduated summa cum laude.

I grew up in the Bay Area, so I’ve been around technology for as long as I can remember. That naturally turned into an interest in computer science and in figuring out how things work behind the scenes. I’m especially interested in distributed systems, databases, backend engineering, machine learning, and AI, but I also enjoy exploring new areas whenever something catches my attention.

Outside of school and work, I’m a big fan of video games and PC hardware. I like keeping up with new CPUs, GPUs, peripherals, and other consumer tech, and I spend a lot of time learning from places like GitHub, YouTube, Stack Overflow, and online developer communities.

I’ve had the chance to work on research, teaching, internships, and a variety of projects, and I enjoy learning from the people I work with just as much as I enjoy building things myself.

education

2026 - 2028

M.S in computer Science

Carnegie Mellon University

PA, USA

Courses: Computer Systems, Algorithms, Distributed Systems, Generative AI

2022 - 2026

B.S in computer Science, minor in Statistics

university of maryland, college park

MD, USA

GPA: 4.0/4.0

GRE: 170Q/161V/5.0W

Relevant Courses: Machine Learning, Multimodal Deep Learning, Database System Architecture and Implementation, Operating Systems, Networks, Design and Analysis of Computer Algorithms, Probability Theory, Theory and Methods of Statistics

skills

Languages: Go, Python, C++, JavaScript, C, Java, SQL

Tools & Frameworks: Git, Linux, AWS, Thrift, Protobuf, GORM, Kitex, Node.js, Express.js, React.js, Docker

AI Systems and Machine Learning: Model Context Protocol (MCP), RAG, PyTorch, LangChain, pandas, scikit-learn, NumPy

Databases: PostgreSQL, MySQL, MongoDB, Qdrant

experience

  • May 2026 - August 2026

    backend software engineer intern

    Global E-Commerce Infrastructure, ByteDance

    bellevue, washington

    Integrated an internal agent platform into Creator Assistant’s Go backend, providing the ReAct agent with configurable skills, MCP tools, and knowledge; built live-data MCPs and optimized tool-schema loading to reduce token usage. Designed and built an agent evaluation harness with automated RPC execution, tool/skill tracing, and blind response review; incorporated real creator conversations alongside curated test cases. Identified an agent observability bottleneck and built a trace viewer that renders model context, tool calls, and internal traces in a readable chat format; developed it into a debugging tool used across the team to diagnose race conditions, tool failures, and response latency. Creator Assistant launched in production for all TikTok Shop Affiliate creators

  • September 2025 - May 2026

    Student Developer

    Data Systems Lab at Maryland (DSLAM), University of Maryland (Dr. Daniel Abadi)

    college park, maryland

    Implemented strict two-phase locking with wait-for graph deadlock detection in a C++ distributed multi-region database. Built distributed transaction validation/commit coordination and asynchronous WAL replication across database nodes.

  • May 2025 - August 2025

    backend software engineer intern

    Global E-Commerce Infrastructure, ByteDance

    bellevue, washington

    Developed the internal Journey Builder UI (React) to streamline configuration of onboarding flows for 250K+ monthly new TikTok Shop creators. Enhanced backend features and built a metrics endpoint (Go, GORM, Kitex, and MySQL). Deployed services via ByteDance’s internal CI/CD pipelines and resolved infrastructure issues with the netlink team.

  • February 2025 - May 2025

    undergraduate student researcher

    cs department, university of maryland (Dr. Deshpande)

    college park, maryland

    Investigated sampling techniques for estimating aggregate queries with expensive predicates evaluated by large language models (LLMs). Designed and benchmarked an importance sampling technique against existing stratified sampling methods.

  • September 2023 - May 2024

    teaching assistant

    cs department, university of maryland

    college park, maryland

    Assist in teaching CMSC216 - Introduction to Computer Systems and CMSC132 - Object-Oriented Programming. Conduct office hours to help students with programming assignments and course topics, proctor exams, and grade assignments.

  • June 2023 - August 2023

    ride operator

    California's Great America

    santa clara, california

    Ensured safe operations of the ride. Enforced ride restrictions and recommendations. Assisted with loading and unloading rides, including small children and guests with disabilities.

  • August 2022 - May 2023

    undergraduate student researcher

    First-Year Innovation and Research Experience (FIRE), University of Maryland

    college park, maryland

    Engaged in Genome Computing research. Used PyMOL, Python, and its libraries, including NumPy, pandas, and Matplotlib, to visualize and analyze DNA-protein interactions.

research & projects


Exploring Importance Sampling for Aggregation on Natural Language Predicates

The project investigated whether importance sampling can outperform uniform and stratified sampling for estimating LLM-based predicate aggregates, ultimately finding that importance sampling is unstable while ABAE remains the most reliable method.
Wang, A. (2024). Exploring Importance Sampling for Aggregation on Natural Language Predicates. Unpublished technical write-up documenting independent research (not peer-reviewed)

UMIACS(University of Maryland Institute for Advanced Computer Studies) Wiki Chatbot

This project built a UMIACS Wiki chatbot that scrapes the entire wiki, embeds and indexes it for retrieval-augmented generation, and fine-tunes an LLM on curated Q/A pairs to deliver accurate, context-grounded answers to user queries.
Wang, A., et al. (2024). UMIACS Wiki Chatbot. Capstone Project Report

Scaling Write Throughput with Multi-Region Validation

Designed and evaluated a multi-region serializable database system implementing 2PL, OCC, and MVCC with remote validation and WAL-based replication.
Nilangekar, P., Wang, A., An, J., Hooker, J., & Zhang, S. (2025). Scaling Write Throughput with a Validation-Based Approach Between Regions. Course project report (equal contribution among authors)

personal projects

portfolio image
Fannie Mae Loan Solution (Bitcamp 2023)
portfolio image
MoneyWatcher: a web application to monitor personal spending
portfolio image
Doodles and Character Classifier
linkedin
linkedin
github
github
AW5@ANDREW.CMU.EDU
AW5@ANDREW.CMU.EDU
website
website