Hello, I'm Yoshi Man.
I build AI-native business intelligence — analytics systems and autonomous agents powered by LLMs.
AI-Native BI Engineer specializing in agentic RAG, autonomous data-analysis agents, and production deployment — bringing AI into how businesses explore and act on their data.
What I'm Building
- Autonomous AI agents with data analytics deep research capabilities
- Agentic RAG production systems for Hong Kong local SMBs
- AI-enabled business intelligence — exploring how AI transforms the future of BI engineering
Background
- 🎓 Master of Data Science — University of Hong Kong
- 🏅 AWS Certified Machine Learning Engineer (2025)
- 💼 5+ years in data & analytics at CHANEL, Under Armour, Flexport
- 🌏 Based in Hong Kong | English, Cantonese, Mandarin
Writing
all →Text-to-SQL Is Harder Than It Looks
Text-to-SQL demos beautifully and breaks quietly. The model nails the syntax — then trips on ambiguity, your schema, and answers that look right but aren't. A field guide to the gap between the demo and a number you'd actually trust.
The Window Is Not the Memory
Retrieval was fixed — then the chatbot got dumber the longer a technician talked to it, and forgot everything by morning. The two things people conflate: the context window and memory.
What Actually Breaks Agentic RAG in Production
I built an agentic RAG system for a cleaning-robot company's field technicians. The agent was the easy part — here's what actually broke in production, and the four fixes that mattered.