👋 I am currently pursuing an M.S. in Electrical and Computer Engineering at the University of Illinois Urbana-Champaign, where I started in Fall 2025 and expect to graduate in May 2027. Prior to this, I earned dual bachelor's degrees in Electrical and Computer Engineering from the University of Illinois Urbana-Champaign and Zhejiang University in 2025. I am actively seeking Ph.D. opportunities in machine learning, large language models, and autonomous agents. Feel free to look at my CV!

My research focuses on how LLMs and autonomous agents can learn and improve from feedback and experience. I am particularly interested in:

• Post-training methods for improving reasoning and decision-making.
• Recursive self-improvement (RSI) that enables agents to learn from multi-step interactions and experimental outcomes.
• Autonomous agents that can advance scientific discovery and solve complex engineering problems.

• University of Illinois Urbana-Champaign
  Aug. 2025 - May 2027 (expected)
  M.S. in Electrical and Computer Engineering (GPA: 3.91/4.00)
  Relevant courses: Reinforcement Learning (ECE586), Natural Language Processing (CS546), Computer Vision (CS543), Graph Machine Learning (CS598).
• Zhejiang University - University of Illinois Urbana-Champaign
  Aug. 2021 - Jun. 2025
  Dual Bachelor's Degrees in Electrical and Computer Engineering (GPA: 3.96/4.00)
  Relevant courses: Computer Systems Engineering (ECE391), Data Structures (CS225), Artificial Intelligence (CS440), Algorithms (CS374), Machine Learning (CS446), Applied Parallel Programming (ECE408).

• 2026 Our paper SAFTAC was accepted to NeurIPS 2026 Main Conference.
• 08/2026 Became a Teaching Assistant for ECE484 (Principles of Safe Autonomy) at UIUC.
• 01/2026 Became a Teaching Assistant for CS446/ECE449 (Machine Learning) at UIUC.
• 08/2025 My paper MAC-Tuning was accepted to EMNLP 2025 Main.
• 08/2025 Started my M.S. in Electrical and Computer Engineering at UIUC.
• 01/2025 Became a Teaching Assistant for MATH285 (Introduction to Differential Equations) at ZJUI.
• 09/2024 Became a Teaching Assistant for MATH241 (Calculus III) at ZJUI.
• 01/2024 Became a Course Assistant for ECE391 (Computer System Engineering) at UIUC.

Please see Research for more information.
(* denotes equal contribution)

• Junsheng Huang*, Yifan Sun*, Zhouer Zhang*, Ning Wei, Yening Liu, Michael Molter, Pavan Kumar Hanumolu, Elyse Rosenbaum, Bin Hu, Huan Zhang. SAFTAC: Simulation-Augmented Fine-Tuning of Open-Source LLMs for Analog Circuit Design. Accepted to NeurIPS 2026 Main Conference.
• Junsheng Huang, Zhitao He, Yuchen Huang, Sandeep Polisetty, Qingyun Wang, Yi R. Fung. MAC-Tuning: LLM Multi-Compositional Problem Reasoning with Enhanced Knowledge Boundary Awareness. Accepted to EMNLP 2025 Main Conference.
• Jason Vega, Junsheng Huang, Gaokai Zhang, Hangoo Kang, Minjia Zhang, Gagandeep Singh. Stochastic Monkeys at Play: Random Augmentations Cheaply Break LLM Safety Alignment. Preprint.
• W. Yan, Z. Deng, M. Molter, R. Krishna, N. Rouf, S. S. Sofla, E. Thieme, S. Chatterjee, P. P. P. Linn, Y. Wen, X. Xiao, Z. Zhang, Junsheng Huang, Y. Liu, Y. Zhou, P. Franzon, M. Swaminathan, E. Rosenbaum, and A. Aysu. From Predict to Generate: A Cross-Layer Survey of Generative AI for IC Design and Manufacturing. Preprint.

I am currently investigating the evaluation and self-improvement of autonomous research agents with advisor Chao Zhang. My research also includes simulation-augmented post-training for analog circuit design, knowledge-boundary awareness, and LLM safety evaluation. Please see my research experience for details.

Please click here for more information.

• ECE484: Principles of Safe Autonomy, Teaching Assistant
  Aug. 2026 - Dec. 2026
• CS446/ECE449: Machine Learning, Teaching Assistant
  Jan. 2026 - May 2026
• MATH285: Introduction to Differential Equations, Teaching Assistant
  Jan. 2025 - Jun. 2025
• MATH241: Calculus III, Teaching Assistant
  Sep. 2024 - Dec. 2024
• ECE391: Computer System Engineering, Course Assistant
  Jan. 2024 - May 2024

Please see Teaching and Leadership for details, including my student representative experience.

• Programming Languages: Python, C/C++, CUDA, x86 Assembly, MATLAB.
• Frameworks & Tools: PyTorch, Hugging Face Transformers, PEFT/LoRA, DeepSpeed/FSDP, vLLM, verl.
• Spoken Languages: English (Proficient), Mandarin (Native), Cantonese (Native).

Feel free to drop me an email if you're interested in collaborating with me. Please see my research for details.

Beyond research, I enjoy basketball, table tennis, billiards, and boxing.