Ian Tu's Website

Cheng-Yang Tu

I'm Engineer

Taking AI products from prototype to paying customers — owning the LLM/RAG pipeline down to on-prem infrastructure.

About

Dual Master’s degree graduate in Computer Science, with degrees from the University of Aizu (Japan) and Chaoyang University of Technology (Taiwan). Specialized in AI model architecture optimization (LLM, MoE) and multi-agent systems, with a blockchain applications background. Currently an AI Engineer at an early-stage hospitality-tech startup, where I took a customer-service chatbot from MVP to a live multi-tenant SaaS product — owning the multi-agent LLM/RAG pipeline, backend, and on-prem infrastructure end to end.

AI Engineer & Blockchain Technology Background.

When dreams merge with action, every breakthrough becomes the beginning of a new world.

Birthday: 20 Aug 2000
Website: chengyangtu.github.io
Phone: (+886) 988-846-121
City: Taichung City, Taiwan

Age: 25
Degree: Master
Email: bravetcy@gmail.com

Resume

Sumary

Cheng-Yang Tu

Open to new opportunities #AI Engineer #LLM Engineer #Multi-Agent Systems #Machine Learning Engineer
Demonstrated leadership as president of a university English speech club and chair of a regional competition.
Life learning, Interdisciplinary passion, Team collaboration

Professional Experience

AI Engineer — Hospitality AI SaaS Startup

Dec 2025 – Present

Early-stage startup, Taiwan

Built the core conversational-AI engine from MVP to production: a multi-agent LLM pipeline (intent classification → identity resolution → task orchestration → booking/RAG sub-agents) serving hotel guests through chat.
Independently re-architected a hardcoded single-tenant prototype into a dynamic multi-tenant SaaS platform, enabling new-tenant onboarding with no code changes.
Cut per-message LLM cost by roughly 60% by designing a tiered, MoE-style model-routing strategy across model sizes, productized into the company's usage-based pricing.
Designed the RAG/vector-search layer and made the architectural call to trade off vector retrieval against deterministic routing for accuracy, latency, and cost.
Owned infrastructure end-to-end: migrated services from cloud to self-hosted on-prem, built a multi-environment deploy pipeline, and load-tested to define capacity and scaling limits.
Delivered the platform's first live paying customer, using real support-conversation data to drive onboarding decisions.

Education

Dual Master’s Degrees in Computer Science

Oct 2023 – Sep 2024 & Sep 2022 – Jun 2023

The University of Aizu (Japan) & Chaoyang University of Technology (Taiwan)

Research Area: Specialized in AI model optimization (LLM, Mixture-of-Experts) (Link)and blockchain technology applications

Bachelor of Computer Science and Information Engineer

Sep 2018 - Jun 2022

Chaoyang University of Technology, Taichung, Taiwan

Activities and Societies: Contest Chair in Speech and Table Topic Contest, President of Chaoyang Tech Toastmasters Club(CTTC), Music Club(Freshman year)，Open Source Tech Club(Junior year)

Publications

Blockchain-enabled transparent traffic enforcement for sustainable road safety in cities

Chin-Ling Chen, Cheng-Yang Tu, Yong-Yuan Deng, Der-Chen Huang, Ling-Chun Liu and Hsing-Chung Chen

Frontiers in Sustainable Cities, 2024 (Paper)

Applications of Computer Vision in Transportation Systems: A Systematic Literature Review

Yu-Kai Lin, Cheng-Yang Tu, Leo Kurosawa, Jia-He Liu, Yu-Zhi Wang and Debopriyo Roy

EDP Sciences, 2024 (Paper)

Honors & Awards

Scholarship for Excellent Students to Study Abroad, Ministry of Education Taiwan, 2023

Third Place in International Symposium on Big Data and ERP Applications, 2023

Technical Skills

LLM / GenAI: Multi-agent pipelines, RAG, prompt engineering, model routing & cost optimization, OpenAI API, pgvector / embeddings
Algorithmic Trading: Strategy development using Pine Script and Power Language, backtesting, and performance evaluation
Machine Learning: Model optimization, Transformer-based models, Mixture-of-Experts (MoE), Text-to-Text generation
Blockchain Development: Smart contract implementation, decentralized applications (DApps), experience with Golang
Backend: Laravel / PHP, PostgreSQL, REST APIs, n8n workflow automation
DevOps: Docker, Nginx, Linux (RHCE), multi-environment deployment, on-prem infrastructure
Linux System Administration: Server configuration, shell scripting, network management, security hardening
Programming Language: Python, C#, C++, HTML, CSS, JavaScript, LaTeX