|
I am a first-year PhD student at Fudan University, supervised by Yinghui Xu, Zenglin Xu and Yu Cheng. My past industry intern experiences includes working on a series of pre-training projects like Step3.5-Flash, Step3, OpenCoder and INF-34B. It's fun to seek the principles underlying the dark magic. Feel free to contact me through email :-) |
|
|
|
|
Step3 & Step3.5-Flash StepFun FoundationModel Team. Step's next generation Foundation Model. ModelCard / Github / Huggingface / Zhihu |
|
OpenCoder: The Open Cookbook for Top-Tier Code Large Language Models Siming Huang*, Tianhao Cheng*, Jason Klein Liu, Jiaran Hao, Liuyihan Song, Yang Xu, J. Yang, J.H. Liu, Chenchen Zhang, Linzheng Chai, Ruifeng Yuan, Zhaoxiang Zhang, Jie Fu, Qian Liu, Ge Zhang, Zili Wang†, Yuan Qi, Yinghui Xu, Wei Chu† ACL2025, ORAL Home Page / TechReport / Github / Huggingface / Zhihu |
|
INF-34B: INF’s Open-Source Large Language Models Jiaran Hao, Zili Wang, Siming Huang, Tianhao Cheng, LiuYihan Song, Ansheng You, Zhipeng Zhou, Xiaoyu Tan, Dakuan Lu, Xiaoming Shi, Chao Qu, Haozhe Wang, Yinghui Xu, Wei Chu, Yuan Qi INF Website Paper / Github / Huggingface |
|
|
|
May 2026 - Present Foudation Model Research Intern I'm exploring agent harnesses and black-box reinforcement learning. |
|
Dec 2024 - Sep 2025 Foudation Model Research Intern As part of pre-training team, I mainly focus on code ability. Due to certain structural factors, Step3's code ability did not achieve the expected results, sadly😿 I met innovative peers, motivated to improve my sense, chasing fundamental issues in LLM. Focus on what matters, and let go of the rest. |
|
Jun 2024 - Nov 2024 Research Intern at FoundationModel Team, supervised by Zili Wang and Jiaran Hao. I'm core author for OpenCoder and main contributor for INF-34B FunFact: I met my current Ph.D. advisor while completing this internship🦁 |
|
July 2023 - Jun 2024 Core Developer for yival (2.5k+ star), 1w+ lines code contribute YiVal is a framework designed for prompt engineering especially in auto-prompting. During my frontend internship, I found prompt engineering too tedious, so I decided to automate prompt design | evol | evaluate. YiVal was my first entrepreneurial venture, but I finally quit due to some ideological differences. |
|
Feb 2023 - Jun 2023 I developed LAB's first LLM inference serving framework based on FasterTransformer. Recognizing the potential of LLMs, I shifted my research focus before beginning my master's. |
|
|
May 2022 - Dec 2022 Develop Tiktok Anchor Recommendation Framework, supervised by Rico Eric My greatest achievement is learning how to solve problems independently. I was happy to find myself growing into a reliable engineer. |
|
|
|
Sep 2025 - Present Ph.D. in Computer Science Member of AI3-Xu Lab, supervised by Yinghui Xu and Zenglin Xu. |
|
|
Sep 2023 - June 2025 Master in Computer Science Member of CMIT, supervised by Prof. Rui Feng. |
|
Sep 2019 - July 2023 B.S. in Computer Science GPA: 3.91/4.00, ranking: 1/163 Member of SHU ACM Training Team, Leader of Well-Prepared Team These are my golden hours |
|
|
|