Yixin Huang

About Me

I am a recent graduate from UC Berkeley with a B.A. in Computer Science and a B.A. in Applied Mathematics (May 2025, GPA: 3.86/4.00). As an AI/ML Engineer and Researcher, I am passionate about advancing the field of AI, particularly in LLM evaluation and MLSys, and driven to create technology that understands and enhances human experiences.

Currently, I am expanding my expertise by learning about MLSys and CUDA to deepen my understanding of system-level optimizations and GPU programming for AI workloads.

Experience

During my studies at UC Berkeley, I had the privilege of working as a Research Assistant at the Sky Computing Lab, advised by Prof. Joey Gonzalez and Prof. Ion Stoica. In this role, I engineered a scalable data pipeline for Openfunctions-v3, led data cleaning efforts for the BFCL-v3 leaderboard (which received recognition from organizations like Meta, Salesforce, and Alibaba)

I also contributed to the Chatbot Arena Project at LMSYS.org, focusing on crowdsourced AI evaluation. I have contributed to enhancing LLM evaluation methodologies using LLM-as-a-Judge for Arena-Hard-v2.

Interests

My primary interests lie in the fundamental challenges of training large-scale models, improving the capabilities and reliability of AI systems, and exploring the frontiers of LLM evaluation. I am dedicated to developing technology that not only performs effectively but also aligns with human values and experiences.

Skills

• AI/ML & Deep Learning: PyTorch, TensorFlow, Hugging Face, LLMs, Transformers, Fine-tuning, RAG, NLP, vLLM, Unsloth
• Programming Languages: Python, C++, Java, Rust, JavaScript, SQL, Bash, C#
• Data Science & Analytics: Scikit-learn, XGBoost, LightGBM, Pandas, NumPy, Spark, MapReduce
• Backend & Infrastructure: FastAPI, PostgreSQL, Docker, RabbitMQ, Linux, Git, RPC

Education

UC Berkeley

• B.A. Computer Science
• B.A. Applied Mathematics
• Graduated: May 2025
• GPA: 3.86/4.00