My Github Home Page

I’m a fourth-year undergraduate student at the University of California, San Diego, majoring in Data Science.

I’m passionate about exploring and analyzing data, building machine learning and deep learning models, and developing full-stack applications powered by large language models or deep learning systems.

Thank you for visiting my website — feel free to explore my projects!

📌

🏢

📧

Research & Internship

Multimodal Misinformation Benchmark Research

Multimodal Learning Video Understanding Machine Learning Benchmarking

Built a full data and training pipeline for a cross-platform health-video misinformation benchmark (YouTube, TikTok, Reddit).
Implemented standardized preprocessing, multimodal feature extraction, and reproducible evaluation workflows.
Conducted systematic experiments across CNN, temporal Transformer, CLIP-style, and LLM-based video models.
Maintained model configurations, experiment scripts, and documentation for consistent cross-architecture comparison.

LLM Reasoning & Interpretability Research

Large Language Models Reasoning Interpretability Representation Analysis

Investigated overthinking and underthinking behaviors in small LLMs through controlled reasoning-length fine-tuning.
Collected and analyzed reasoning trajectories from closed-source models for mechanistic comparison.
Studied shifts in residual streams and concept activations across different reasoning regimes.
Designed experiments to probe reasoning stability beyond heuristic prompting techniques.

Kami Vision — LLM Application & Multimodal System Engineering Internship

Large Language Models RAG Backend Engineering Multimodal AI

Developed multimodal LLM-based video monitoring features, including event summarization and multilingual Q&A.
Designed and implemented a production RAG pipeline with hybrid retrieval and FAISS-based vector alignment.
Diagnosed and resolved inference bottlenecks across APIs, model servers, and cloud-integrated services.
Built a scalable backend (parallel APIs, streaming response) powering real-time video intelligence applications.

Resume

Download Resume

Hi, I am georgong

My Projects

Research & Internship

Multimodal Misinformation Benchmark Research

LLM Reasoning & Interpretability Research

Kami Vision — LLM Application & Multimodal System Engineering Internship

Resume