Rongyuan Tan is a Ph.D. Candidate in Computer Science and Technology at Southern University of Science and Technology (SUSTech), Shenzhen, China. His research interests include cloud computing, distributed systems, microservice systems, AIOps, and LLM interpretability.

Contact

• Email: 12231141@mail.sustech.edu.cn
• Phone: +86 15895640109
• Location: SUSTech, Shenzhen, China

Education

• 2022–Present — Ph.D., Computer Science and Technology, Southern University of Science and Technology, Shenzhen, China
  Relevant courses: Cloud Computing, Distributed System, Microservices System.

• 2019–2021 — M.Res., Computer Science, University of Liverpool & Xi’an Jiaotong-Liverpool University, Suzhou, China
  Relevant courses: Human-Computer Interaction, Big Data Analysis, Cloud Computing.

• 2015–2019 — B.Sc., Computer Science and Technology, Shenyang Jianzhu University, Shenyang, China
  Relevant courses: Software and Hardware, Data Visualization, Software Engineering, Algorithms, Game Designing.

Academic Publications and Research

• MAAD: A Distributed Anomaly Detection Architecture for Microservices Systems (IPDPS 2024)
  Proposed a multi-agent distributed anomaly detection system for microservices, addressing the limitations of centralized, large, GPU-dependent models and low-accuracy lightweight methods. The framework outperforms baseline methods in detection accuracy, efficiency, and overhead.

• Congestion-aware Microservice Scheduling Framework Based on Hybrid Scheduling Strategies (under review)
  Designed a hybrid scheduling framework combining a rapid scaler for sudden traffic/QoS violations and a class decision tree-based strategy fusion model for long-term resource optimization.

• Distributed Fault Diagnosis Architecture Based on Multimodal Data for Microservice Systems (submitted to IEEE Transactions on Service Computing)
  Built a multi-agent diagnosis system over traces, logs, and metrics for anomaly detection and service-level root cause localization, achieving strong Top-1/Top-5 accuracy with low latency.

• Contrastive Attribution on Realistic Benchmarks (under review, Microsoft internship work)
  Studied contrastive LRP-based attribution for LLM failure debugging, and proposed an efficient method to construct cross-layer attribution graphs in long-context settings.

Work Experience

Research and internship experience in industry collaborations and applied AI systems:

• 07/2025–12/2025 — Research Intern, DKI Group, Microsoft, Shanghai, China
  Conducted joint research with Microsoft on AIOps and LLM interpretability, leading to peer-reviewed publications.

• 05/2021–09/2021 — Computer Vision Algorithm Intern, XiaoIce, Microsoft, Suzhou, China
  Tested and optimized deep learning algorithms including pose estimation, face warp, and music2dance; strengthened Linux, Python, and C++ engineering skills.

• 07/2020–08/2020 — Game Development Intern, Seasun Game (Kingsoft Group), Zhuhai, China
  Co-developed a 3D parkour game in Unity, responsible for project management and player presentation logic implementation.

Honors and Awards

• Second Scholarship of Excellent Student (Top 3%–10%) — 1 time
• Learning Excellence Scholarship — 1 time
• Volunteering Advanced Individual — 2 times
• Third-class Scholarships of Excellent Student (Top 10%–30%) — 5 times
• Social Work Scholarships — 5 times
• Advanced Individual in Social Practice during Holiday — 2 times
• Excellent Student Cadres — 5 times

Skills

• Cloud Computing: Kubernetes, Docker

Languages

• Chinese: Native
• English: Skilled