resume

General Information

Full Name	Albert (Geyang) Xu
Email	gxu048@ucr.edu
Phone	+1 858-568-6771
Website	albertgy9910.github.io
Location	Riverside, CA

Research Interests

Testing & reliability of deep learning compilers
Reliability of AI systems
Software engineering for machine learning
Data-centric robustness of ML pipelines

Education

2026.09 - 2029.06 (expected)
Ph.D. in Computer Science

University of California, Riverside
- Software Engineering for Machine Learning
- Advisor: Prof. Qian Zhang
2022.09 - 2024.03
Master of Science in Computer Science

University of California, San Diego
- GPA: 3.87/4.0
2018.09 - 2022.06
Bachelor of Science in Computer Science

University of Liverpool, UK
- GPA: 3.93/4.0 (top 3%)

Experience

2025.10 - 2026.04
Manufacturing Data Analyst

Fuyao Glass America Inc.
- Built Python/SQL pipelines that fuse machine and inspection-system data with coating and tempering signals.
- Productionized the pipelines as cron jobs with alerts and access control: 99.3% job success, 2.4 hrs/week less downtime.
- Shipped role-based dashboards tracking OEE, FPY, scrap, and changeover; ran A/B and DOE setpoint studies for process tuning.
2024.10 - 2025.10
Back-end Engineer

4Pexonic Inc.
- Developed an academic impact analysis platform mining citation patterns across tens of thousands of records using Node.js, MongoDB, and Next.js with Docker, serving university researchers and policy institutions.
- Built two-tier caching (in-process LRU + Redis), cutting median and p95 latency by 40%.
- Integrated gRPC microservices; implemented bcrypt/session auth and Nginx TLS.
2023.10 - 2024.10
Research Engineer

UCSD Halıcıoğlu Data Science Institute
- Built the error-injection pipeline of a black-box ML stress tester (beam search + Optuna Bayesian optimization).
- Benchmarked 8 data-cleaning, 4 debiasing, and 2 uncertainty-quantification methods; at 10% corruption the framework cut AUC to 0.58 vs. 0.79–0.85 for random corruption and gradient-based poisoning.

Publications

2025
Stress-Testing ML Pipelines with Adversarial Data Corruption

VLDB 2026, PVLDB 18(11)
- J. Zhu, G. Xu, et al.
- DOI: https://doi.org/10.14778/3749646.3749721
- Code: https://github.com/lodino/savage

Skills

Languages
- Python, Java, Go, SQL, C/C++, JavaScript, R
Databases
- PostgreSQL (pgvector), MongoDB, Redis
Frameworks and Tools
- Docker, Git, gRPC, Node.js, Next.js, Django, React, Optuna, Jupyter, AWS, LangGraph