About Me
I am a 1st-year PhD student at Chungbuk National University (CBNU), working in the Big Data Lab under the supervision of Professor Wanseop Cho (์กฐ์์ญ). My research interests focus on Reinforcement Learning and Large Language Models, particularly LLM reasoning, Chain-of-Thought (CoT) optimization, and reinforcement learning approaches for improving the efficiency and reliability of AI reasoning.
Research Interests
Languages
- Python
- C++
- Java
Frameworks
- PyTorch
- TensorFlow
- Hugging Face
- FastAPI
Tools
- Docker
- Git
- LaTeX
- Linux
- VLLM
Experience & Education
PhD in Computer Science (In Progress)
Chungbuk National University
|Cheongju, South Korea
Mar 2026 - Present
Research focus: Reinforcement Learning, LLMs, RAG, CoT. Member of the Big Data Lab.
M.S. in Big Data
Chungbuk National University
|Cheongju, South Korea
Sep 2023 - Aug 2025
Research focus: machine learning, Information Retrieval, RAG, Domain Adaptation, and LLMs. Member of the Big Data Lab.
B.S. in Information Technology
Simad University
|Mogadishu, Somalia
2013 - 2018
Foundation in algorithms, data structures, and software engineering.
Research Experience
Big Data Lab Researcher
Chungbuk National University
|Cheongju, South Korea
Sep 2023 - Present
Conducting research on RAG and large language models. Developing agenitc AI, Domain Adaptation, and RAG-based systems.
Industry Experience
App Developer
SOSTEC
|Mogadishu, Somalia
May 2018 - Aug 2022
Developed production mobile applications using React Native and backend services with Laravel. Led cross-functional teams to deliver high-quality software solutions.
Research & Projects
LLM Reasoning & Explainability: Thinker-Judge
Developed a dual-model inference framework for transparent LLM reasoning, combining a thinking model for multi-step reasoning with an LLM-based judge for evaluating logical consistency and factual grounding.
ARC-AGI: Improving LLM Reasoning
Researching how to improve LLM reasoning capabilities using the ARC-AGI benchmark to evaluate abstract reasoning, few-shot generalization, and rule induction from minimal examples.
Multi-Agent AI Framework
Developed a multi-agent AI system for domain-specific question answering, combining intent-based agent routing, Structure-Aware RAG, FAISS-GPU retrieval, and RAGAS-based evaluation.
Publications
No publications available at the moment.
Awards & Honors
Global Korea Scholarship (GKS) for Graduate Studies
Chungbuk National University (CBNU) & | 2022 - 2025
Get In Touch
I am always open to discussing research collaborations, potential projects, and new ideas. If you are interested in working together or would like to learn more about my research, I would be happy to hear from you.
Contact me
ugasbanin@gmail.com
Institution
Chungbuk National University
Location
Cheongju, South Korea