I am Kevin Ray Scott (雷凯文 / LEI KAIWEN), a software engineering and artificial intelligence student working at the intersection of language models, robotics interaction, data analytics, computer vision, and applied machine learning. I am especially interested in AI systems that are useful outside a controlled notebook: systems that can retrieve evidence, explain where information comes from, handle failure states clearly, and support practical human workflows.
My academic background combines undergraduate study in Software Engineering at Harbin Normal University with postgraduate study in Artificial Intelligence at Universiti Malaya. My recent work includes LLM fine-tuning with SFT/LoRA, retrieval-augmented generation, agent harness engineering, ROS-based smart-home interaction, temporal gesture recognition, climate analytics, multimodal AI, and medical AI experimentation.
I like building the full path from data preparation and model adaptation to evaluation, interface design, and deployment-ready workflows. Research-wise, I am drawn to reliable AI, evidence-grounded generation, controllable agent behavior, explainable evaluation, and human-centered intelligent interfaces. I am looking for opportunities where I can contribute as an engineer while continuing to grow toward deeper AI research, including future PhD-oriented work.