Hovhannes Grigoryan
Causal identification and mechanism design for agent-driven decision systems.
Senior Applied Scientist at Amazon (NYC) · Physics PhD · former quantitative analyst. I study causal methodology, mechanism design, reliable AI agents, and their intersections, and write about each.
Writing
No long-form posts yet. First two posts, research agenda and physics-to-applied-AI transition, coming soon.
Lectures
A working textbook on causal inference, agent-driven decision systems, and the mathematical foundations of intelligence. Full chapters with learning outcomes, derivations, numbered equations, exercises, and bibliography, rendered in academic-textbook style for seminar use.
Read the first chapter: Double Machine Learning →
Research notes
A working library on the methodological foundations of causal identification, agents and reliability, the contemporary LLM frontier, and quantum computing. Twenty-eight research notes across five thematic groups.
Applications
Compact demos of analytical methods applied to financial and statistical problems. Public data or synthetic DGPs only, reproducible figures, documented limitations, no Amazon content.
About
Physics PhD (2008) and six years of postdoctoral research; quantitative analyst at Citi Global Markets (MBS); Quantopian Trading Contest winner (4 months, 2019); five years of adjunct teaching at CUNY, Manhattan College, Georgetown; Amazon Senior Applied Scientist since August 2024.
Contact
- Email,
iohanngrig@gmail.com - GitHub,
github.com/iohanngrig - Hugging Face,
huggingface.co/iohanngrig - LinkedIn, linkedin.com/in/iohanngrig