About Me
Welcome to my blog! I’m Yuchen Liu, a Machine Learning Engineer at Amazon Japan. I identify as an AI Product Engineer — I build AI-native products end-to-end using agentic coding.
My work centers on context engineering for AI agents: designing structured access to retail business data, science models, and domain knowledge so that agents can reason and act effectively. I build MCP servers that equip business roles with data analysis, knowledge retrieval, and workflow automation capabilities. I also develop evaluation frameworks to measure whether agents use context correctly — because shipping an agent is one thing; knowing it actually works is another.
Beyond my day job, I’m passionate about the frontier of LLMs and agentic systems. I actively experiment with foundational models, agentic framework, and enterprise agent ecosystem — both inside Amazon and in personal projects — pushing the boundaries of what agentic AI can deliver in production.
I also run a video channel on BiliBili, originally launched during my graduate school years at Tokyo Institute of Technology, where I shared experiences about studying and living in Japan. Moving forward, I plan to expand this blog to focus entirely on technical content: deep dives into AI agents, MCP design patterns, practical tutorials, and innovative project ideas at the intersection of LLMs and real-world applications.
I’m open to consulting and collaboration opportunities, particularly around AI agent architecture, context engineering, and LLM application development. If you’re working on something interesting, feel free to reach out via email or SNS.
Thanks for stopping by — I hope you find something useful here.