M.S. in Electronic Information (Key Software)
Peking University, School of Software & Microelectronics
About
I am an M.S. student at Peking University, with a background in computer science from Nanjing University. My work and project experience span large language model systems, generative retrieval and recommendation, efficient model deployment, and multi-agent reinforcement learning. I am particularly interested in building reliable and efficient AI systems that connect learning algorithms with real-world infrastructure.
Research interests: LLM systems, generative retrieval, recommender systems, model inference and deployment, multi-agent reinforcement learning.
Education
B.S. in Computer Science and Technology
Nanjing University, Kuang Yaming Honors School
Experience & Projects
TikTok Search — Generative Retrieval for Short-Text Recommendation
ByteDance · Architecture Intern, Search
- Contributed to a unified generative retrieval pipeline for short-text recommendation and the deployment of OneSug-based generative recommendation techniques.
- Improved serving latency and resource utilization across the recommendation pipeline.
- Collaborated with the XLLM inference framework team on model deployment and performance optimization for different generative serving workflows.
DevMate — Domain-Specific Large Language Model for Data Communications
Huawei Nanjing Research Institute · Software Development Intern
- Participated in the design of a multi-agent LLM system for DevMate.
- Integrated knowledge retrieval from the company product data lake, designed and implemented an intermediary API, and curated training corpora for model fine-tuning.
- Trained and deployed a TinyBERT classifier on Ascend NPUs with MindSpore for domain classification of user queries.
- Represented the department in Huawei's Hackathon Challenge, advanced to the final as a top-three team from the Nanjing Research Institute, and received third prize.
Multi-Agent Reinforcement Learning & Decision Foundation Models
LAMDA, Nanjing University · Research Intern
- Studied hierarchical multi-agent reinforcement learning and decision foundation models.
- Contributed to the early stage of a project on language-guided multi-agent reinforcement learning with hierarchical policies and human feedback.
- Conducted literature review, reproduced related methods, and participated in model design and implementation.
- Explored a framework in which agents first learn low-level policies and then use RLHF to follow natural-language instructions for high-level decision making.
Selected Honors
Programming Contests
Selected competitive programming awards
- 2nd CCF Algorithm Capability Competition (CACC) — Gold Medal4th · Apr 2026
- 46th ICPC Asia Regional, Kunming — Silver Medal36th · Apr 2022
- Jiangsu Collegiate Programming Contest — Gold Medal13th, unofficial · May 2023
- 9th CCPC Regional, Shenzhen — Silver Medal67th · Nov 2023
- 47th ICPC Asia Regional, Hefei — Silver Medal73rd · Nov 2023
- 7th CCPC Regional, Guilin — Silver Medal65th · Nov 2021
Scholarships & Distinctions
Nanjing University
- Outstanding Graduate, 2025
- Special Prize, Basic Sciences Scholarship, 2023
- China Merchants Bank All-in-One Card Scholarship, 2022
- Excellence Award, Basic Sciences Scholarship, 2022 and 2024
- Yang Yongman Scholarship, 2021 and 2023
- Outstanding Student, Freshman College, 2022
Languages
- Chinese: Native
- English: CET-6 640; TOEFL 93
Contact
The best way to reach me is by email at litrehinn@gmail.com. You can also find my public projects on GitHub.