Welcome to my GitHub! I'm Seyeong Han, currently pursuing my Master's in Engineering Management at the University of Texas at Austin, with a strong background in machine learning, deep learning, and AI-driven solutions. My passion lies in leveraging cutting-edge technology to solve real-world problems, whether it's optimizing AI models for speed and efficiency or developing innovative solutions to address unique challenges.
I have over three years of experience as a Machine Learning Engineer, with a track record of developing and deploying AI solutions across various industries. As the CTO and Co-founder of Resia, an AI-driven home renovation platform, I successfully integrated AWS services to enhance scalability and security, securing $8,500 in funding to accelerate our vision.
My previous roles at Nearthlab and SNUAILAB involved building MLOps pipelines, developing custom models for wind turbine crack detection, and optimizing AI models to achieve significant performance improvements. I also have a strong foundation in deep learning frameworks like PyTorch, TensorFlow, and NVIDIA’s AI tools, as well as experience in DevOps and cloud services.
- Deep Learning: PyTorch, TensorFlow, NVIDIA Deepstream, TensorRT, HuggingFace, ONNX
- DevOps: AWS (Sagemaker, Bedrock, Amplify), Docker, Jenkins, Airflow, Grafana
- Programming: Python, CUDA, JavaScript, C++, Java, ReactJS, Vue.js
- Tools & Libraries: Neo4j, MLFlow, OpenCV, Open-MMLab
- LlamaIndex Webinar
- Github
- Original Repo
As the technical lead, I spearheaded the development of "memary," a personalized Retrieval-Augmented Generation (RAG) model, in collaboration with LlamaIndex. This open-source project, which achieved over 1K stars on GitHub, was designed to enhance memory capabilities in AI by integrating graph storage and RAG techniques.