Jiacheng Miao

I am a Data Science Postdoc Fellow at Stanford University, working with Jonathan Pritchard and James Zou. Things I'm excited about:

• How to build and train Agentic AI systems that augment and accelerate research?
• What capabilities are current LLMs missing for doing research?
• Can we simulate researchers with AI?

Previously, I received my PhD at the University of Wisconsin–Madison, advised by Qiongshi Lu and Lauren Schmitz. I did my BS in Department of Mathematics at Nanjing University. During my PhD, I developed methods and theory for reliable use of AI/ML-generated data in scientific analyses and for understanding how genes and environment jointly shape human health.

Email me at jcmiao[at]stanford.edu. See my CV and Google Scholar.

Research  /  Data  /  Software  /  Github  /  Twitter

---

Selected Papers

• Paper2Agent: Reimagining Research Papers As Interactive and Reliable AI Agents [code]
  Miao J., Davis J., Pritchard J., Zou J. 2025. (>2k Github stars)
• Understanding nature and nurture: Statistical and AI innovations uncover how genes and environment shape human health
  Miao J. Science, 2025.
• Valid inference for machine learning-assisted genome-wide association studies
  Miao J., Wu Y., Sun Z., Miao X., Lu T., Zhao J., Lu Q. Nature Genetics, 2024.
• Assumption-lean and data-adaptive post-prediction inference
  Miao J.*, Miao X.*, Wu Y., Zhao J., Lu Q. Journal of Machine Learning Research, 2025.
• Task-agnostic machine-learning-assisted inference
  Miao J. and Lu Q. NeurIPS, 2024.
• PIGEON: a statistical framework for estimating gene-environment interaction for polygenic traits
  Miao J., Song G., Wu Y., et al. Nature Human Behaviour, 2025.