# Yixin Tian

> Yixin ("Isshin", 一心) Tian is a machine learning engineer and data scientist based in Toronto, Canada. He specializes in designing human-centred AI/ML systems at the intersection of software engineering and machine learning research, with expertise in educational psychology, learning science, agentic AI, RAG, knowledge graphs, and MLOps.

## Professional Background

Yixin currently works as a Data Scientist at RBC, developing agentic AI assistants for the credit domain. His experience includes deploying the Dagster orchestration platform, building event-driven inference and monitoring pipelines, and co-developing a patent-pending agriculture carbon-emission model. He served as a team lead for RBC Amplify in summer 2025, mentoring a team to prototype an AI assistant using RAG, text-to-SQL LLM agents, and self-supervised client insight generation algorithms. He is also a part-time researcher at the University of Southern California's Information Sciences Institute (ISI), researching agentic structured data extraction from long documents and knowledge graphs.

Research interests include multi-modal reasoning and planning, graph machine learning, knowledge graphs, neuro-symbolic AI, cognitive science, and educational psychology. Outside of work, he enjoys running, writing, reading, and hiking.

## Featured Blog Posts

- [All blog posts](https://www.yixtian.com/blog): Complete archive of technical articles and writings
- [When knowledge becomes worthless](https://www.yixtian.com/blog/12-when-knowledge-becomes-worthless): AI is making answers cheap, shifting the scarce resource to questions. But questioning ability comes from struggling to find answers—creating a paradox where AI may erode the very skill it makes valuable.
- [The hitchhiker's guide to AI coding tools](https://www.yixtian.com/blog/11-ai-coding-agents-guide): My personal workflow and useful tips for maximizing your leverage when coding with AI copilots/IDEs like Claude Code, Cursor, etc.
- [Multi-agent debate with state pattern from scratch](https://www.yixtian.com/blog/10-multi-agent-debate-w-state-pattern): Building a multi-agent debate framework using LLMs with proponent, opponent, and neutral agents using the state pattern
- [Nexus book review](https://www.yixtian.com/blog/9-nexus-book-review): Review and insights on information networks and knowledge sharing
- [Introduction to Ontology](https://www.yixtian.com/blog/8-ontology-101): Deep dive into ontology concepts for scalable knowledge sharing and the Web Ontology Language (OWL)
- [Three dimensions of learning](https://www.yixtian.com/blog/7-three-dimensions-of-learning): Exploring the three dimensions of learning: cognitive, social-emotional, and metacognitive
- [Blog migration journey](https://www.yixtian.com/blog/6-blog-migration): Technical details about migrating the blog to Next.js with MDX
- [Evolution meets art](https://www.yixtian.com/blog/5-evolution-meets-art): Intersection of evolutionary algorithms and creative expression
- [Modern note-taking apps](https://www.yixtian.com/blog/4-modern-note-taking-apps): Analysis of modern note-taking applications and knowledge management systems
- [Linear regression fundamentals](https://www.yixtian.com/blog/3-linear-regression): Educational content on machine learning basics
- [Question answering over multiple documents using LLM](https://www.yixtian.com/blog/2-rag-w-langchain): Building a semantic QA interface over knowledge-base notes with LangChain, OpenAI GPT, and Chroma vector DB
- [First look at learning](https://www.yixtian.com/blog/1-first-look-at-learning): Initial explorations into learning theory and cognitive science

## Technical Expertise

- **AI/ML Engineering**: Agentic AI systems, RAG (Retrieval-Augmented Generation), text-to-SQL agents, MLOps pipelines
- **Knowledge Representation**: Knowledge graphs, ontologies, semantic web technologies, OWL
- **Machine Learning Research**: Multi-modal reasoning, graph machine learning, neuro-symbolic AI
- **Software Engineering**: Event-driven architectures, orchestration platforms (Dagster), Python development

## Contact & Links

- **LinkedIn**: https://www.linkedin.com/in/yixintian/
- **Personal Portfolio**: https://www.yixtian.com/
- **GitHub**: https://github.com/yixin0829
- **Twitter/X**: @Isshin0829
- **Location**: Toronto, Canada
- **RSS Feed**: https://www.yixtian.com/rss