Education

2024 - present, M.S. in Computer Technology, Institute of Computing Technology, Chinese Academy of Sciences (ICT, CAS)
Admitted with exemption from the national graduate entrance examination (Ranked 1st in class, 3th in major)

2020 - 2024, B.S. in Computer Science and Technology, China University of Petroleum (East China)
GPA: 4.12/5 | Rank: 3/114 (Top 3%)

Internships

Microsoft (Asia) Internet Engineering Institute

Engineering Intern · Multi-Agent AI Systems

Oct. 2025 – Dec. 2025

AutoGen-Based Multi-Agent Diagnostic System

Engineered and optimized a multi-agent diagnostic system for automated error-log analysis, root-cause diagnosis, and retrieval-augmented solution generation, with integration into a Microsoft Teams bot.

AutoGen Multi-Agent Systems Retrieval-Augmented Generation Azure AI Search Microsoft Teams LLM Engineering
  • Designed an AutoGen-based multi-agent diagnostic workflow with specialized agents, tool-driven interaction, dynamic follow-up, and context backtracking for diagnosing complex distributed-job failures.
  • Developed an intelligent error-log processing pipeline combining initial error localization, worker-level log routing, dynamic context expansion, noise filtering, hash-based deduplication, and structured log retention for subsequent reuse.
  • Built a hybrid retrieval and solution-generation pipeline using BM25 and vector search, candidate re-embedding, MMR reranking, dynamic similarity filtering, and BREW semantic compression to retrieve diverse and relevant historical solutions.
  • Optimized system efficiency and bot integration, reducing per-task context from over 20K tokens to approximately 2K and runtime from about 240 seconds to 50 seconds, while integrating the diagnostic workflow into Dri-Bot for Microsoft Teams.
Reported Impact 20K+ → ~2K tokens ~240 s → ~50 s runtime Evaluation: 38.6% → 85.71%

China Mobile

Research Intern · AIGC Video Detection

Jan. 2026 – Mar. 2026

Generalizable Detection of AI-Generated Videos

Conducted research on generalizable AIGC video forensics, focusing on detection under unseen-generator and data-scarce settings, with consideration of robustness and lightweight deployment.

Video Forensics Zero/Few-Shot Learning Diffusion Models Meta-Learning Knowledge Distillation
  • Surveyed modern AIGC video generation and detection methods, with an emphasis on cross-generator generalization, robustness to compression and transcoding, and inference efficiency.
  • Proposed an unknown-generator detection framework that maps videos into a diffusion reconstruction-error space, aligns cross-generator representations through supervised contrastive learning, and distills the teacher model into a lightweight spatiotemporal detector.
  • Designed a complementary zero/few-shot detection scheme combining frequency-domain disentanglement, episodic prototype learning, and one-class anomaly detection for rapid adaptation with few or no synthetic samples.
Deliverables 2 technical surveys 1 model-design proposal

Honors & Awards