Zichen "Charlie" Zhang

Zichen "Charlie" Zhang

AI Lab @ Supercell | Co-Founder & CTO @ Collage

University of Michigan

About Me 👋

I’m a third-year undergraduate student at the University of Michigan, majoring in Computer Science. I’m affiliated to the Honors Program at the College of Literature, Science, and the Arts (LSA).

I’m currently working in the AI Innovation Lab at Supercell, maker of Brawl Stars, Clash Royale, Clash of Clans, Boom Beach, etc.

I’m passionate about training unified foundation models that accept multimodal inputs, including touch, images, videos, language, and audio. I’m enthusiastic about building efficient multimodal models by efficiently fine-tuning pre-trained unimodal models using a limited amount of downstream data and improving unimodal accuracy by transferring representations learned from other modalities.

I’ve designed efficient algorithms that handle large amounts of data, including the MIA-Sort algorithm, and I was fortunate to be advised by Professor Minji Kim at the Minji Lab. I also implemented a study to find factors influencing Brain-Computer Interface (BCI) system’s performance on people with physical impairments, and I was honored to be advised by Dr. Jane Huggins at the Direct Brain Interface (UM-DBI) Laboratory.

I’m also excited about building next-generation AI applications to tackle real-world problems. I co-founded Collage, an EdTech startup. We aim to replace archaic college course registration systems with an AI-powered discovery engine that automates and personalizes college academic advising and scheduling, connecting students with their classes, advisors, and peers. Sign up for free at joincollage.com with Google Education account and start building your perfect college experience.

🔗 You can find my social media accounts here at my Linktree.

📢 Feel free to drop me an email at zhangzzc@umich.edu. I’m always open to chats and feedbacks.

🎨 View my Chinese calligraphy and photography works here!

Interests
  • Unified foundation models
  • Efficiently fine-tuning using little downstream data
  • Transfer learning from other modalities
Education
  • B.S. in Computer Science, Present

    University of Michigan

Research Projects

* denotes equal contribution

*
Learning Pushing Dynamics for Arbitrary 2D Rigid Bodies

Art Boyarov*, Zichen Zhang*.

We study the problem of learning the pushing dynamics of arbitrary 2D rigid bodies, developing neural network models trained on simulated data collected with a Franka Panda robot. By comparing a shallow MLP to a deeper point-cloud-inspired network, we show that the deeper model better captures the complex motion dynamics of different 2D shapes. Using the learned models in a Model Predictive Path Integral (MPPI) controller, we successfully achieve closed-loop pushing and obstacle avoidance across diverse 2D rigid bodies.
Learning Pushing Dynamics for Arbitrary 2D Rigid Bodies
Simpler is Better: Finding the Best Reward Function in Long Chain-of-Thought Reinforcement Learning for Small Language Models

Luning Wang*, Zichen Zhang*, Junkuan Liu*.

We study three types of reward functions — normal, cosine, and dynamic — for long chain-of-thought reinforcement learning in Small Language Models, and find that the simple normal reward consistently outperforms more complex designs, suggesting that simpler rewards are good enough for eliciting reasoning in smaller models.
<em>Simpler is Better:</em> Finding the Best Reward Function in Long Chain-of-Thought Reinforcement Learning for Small Language Models
VTMo: Unified Visuo-Tactile Transformer Encoder with Mixture-of-Modality-Experts

Zichen Zhang, Peihao Li, Yuan Cheng.

We introduce VTMo, a modular Vision-Touch Transformer encoder that unifies dual-encoder flexibility with fusion-encoder accuracy through a shared self-attention mechanism and modality-specific or cross-modal experts. VTMo supports image-only, touch-only, and vision-touch fusion tasks, offering versatility for speed or accuracy. Our method achieves competitive performance on the Image-to-Touch Retrieval task while reducing training time and computational complexity.
VTMo: Unified Visuo-Tactile Transformer Encoder with Mixture-of-Modality-Experts
Babysitting a Small Language Model through One-Step Tree-of-Thoughts Knowledge Distillation

Anurag Renduchintala*, Adi Mahesh*, Zichen Zhang*, Zimo Si*, Shangjun Meng*, Samuel Fang*.

We introduce the One-Step Tree-of-Thoughts framework, a simplified prompting method that distills multi-step reasoning into a single structured prompt, and demonstrates how knowledge distillation can transfer this reasoning capability from Large Language Models to Small Language Models with much less parameters, enabling significant improvements reasoning performance, beating GPT-4o and GPT-4, as shown on the model's performance on Game of 24.
Babysitting a Small Language Model through One-Step Tree-of-Thoughts Knowledge Distillation
MIA-Sort: Multiplex Chromatin Interaction Analysis by Efficiently Sorting Chromatin Complexes

Zichen Zhang, Minji Kim.

MIA-Sort is a Python bioinformatics tool for efficiently extracting and sorting chromatin complexes from large datasets like Hi-C and Pore-C, enabling researchers to analyze chromatin loops, stripes, jets, and hubs to study loop extrusion.
MIA-Sort: Multiplex Chromatin Interaction Analysis by Efficiently Sorting Chromatin Complexes

Engineering Projects

Builds & Products (* denotes equal contribution)

*
GenHint: An AI Coding Assistant that Teaches How to Code
🏆 Honorable Mention Best Developer Tool 🏆. MHacks 17 Hackathon, 2024

Zichen Zhang*, Peihao Li*, Tongyuan Miao*, Yuchen Huang*.

Unlike conventional AI coding assistants like GitHub Copilot, GenHint does not give you codes directly. Instead, it generates code templates with "TODO" comments and explains each subproblem for you. It's powered by Llama 3 70B hosted on Groq. Published as a VS Code Extension and easy to download, it will be a great companion for your coding workflow.
GenHint: An AI Coding Assistant that Teaches How to Code
Collage: Pinterest for Advising
🌟 500+ Active Users & 300+ Mentor Queries since launch. 🌟

Max Feldman*, Alex Wang*, Zichen Zhang*, Jaden Sun*, Kali Francisco, Nate Bennett, Leo Choi, Tanishka Nalawade, Joe Tang, Asher Katz, Eden McCullough, Dasha Skalitzky.

Accessible via your Google Education account, Collage is an AI-powered educational discovery engine that personalizes academic advising and scheduling and connects students with their classes, advisors, and peers. The platform features a personalized, Pinterest-like course catalog tailored to individual academic interests and career goals, an AI advisor to assist with scheduling decisions, and a social networking system for sharing schedules and connecting with peers.
Collage: Pinterest for Advising

Experience

 
 
 
 
 
Supercell
AI Innovation Lab
April 2025 – Present Helsinki, Uusimaa, Finland
Integration of Multimodal LLMs into games to create richer, more immersive end-user experiences
 
 
 
 
 
Collage
Co-Founder & CTO
April 2024 – Present Ann Arbor, MI, USA
Nuking archaic course registration systems with web search & text-to-SQL LLM agent, Retrieval-Augmented Generation, and recommendation algorithms
 
 
 
 
 
Starward Game Studios
AI Engineering Intern
April 2025 – Present Mountain View, CA, USA
LLM Continued Learning, Retrieval-Augmented Generation, and agents in game characters
 
 
 
 
 
U-M Minji Lab
Research Intern
May 2024 – December 2024 Ann Arbor, MI, USA
MIA-Sort, first bioinformatics Python package that efficiently reads datasets of 4B+ chromosome fragments and sorts/visualizes them in different schemes, open-sourced on GitHub
 
 
 
 
 
Michigan Hackers
Core Member
September 2022 – May 2023 Ann Arbor, MI, USA
MWalk, an Android mobile app written in Java that tracks U-M students’ walking workouts, speed and distance, using Health Platform API and Firebase
 
 
 
 
 
U-M Direct Brain Interface Laboratory
Research Intern
September 2022 – May 2023 Ann Arbor, MI, USA
User interaction automation of an internal Brain-Computer Interface (BCI) survey research instrument

Awards

MHacks 2024
Honorable Mention Best Developer Tool
One of the largest hackathons in the U.S., attracting over 550 students from leading universities in North America.
See certificate
University of Michigan
University Honors
Received for 5 terms: Fall 2022, Winter 2023, Fall 2023, Winter 2024, Fall 2024
See certificate
University of Michigan
James B. Angell Scholar
“A” record for two or more consecutive terms. Received for two years, 2024 and 2025
See certificate
University of Michigan
William J. Branstrom Freshman Prize
First-term freshmen who rank in the upper 5%
See certificate

Fun Facts

  • My Chinese name is 张紫宸. In Chinese, 'Zi' (紫) means 'purple,' which traditionally represents nobility, elegance, and auspiciousness. 'Chen' (宸) refers to a palace or the residence of an emperor, symbolizing dignity and grandeur. Together, the name 'Zichen' conveys a sense of royalty, grace, and aspiration for greatness.

  • I was born and raised in Zhangjiagang, a small city close to Suzhou, China. I studied at the Jiangsu Provincial Liangfeng Senior Middle School (梁丰高中) International Department (now named International Academy) in China before transferring to Rochester Adams High School in Rochester Hills, Michigan. Go Highlanders!

  • I love Chinese calligraphy and I started learning it since elementary school. Please follow me on Instagram to see more of my works!

Photography

Capture Moments in Life