Michael J. Ryan

Michael J. Ryan

PhD student in NLP

Stanford University

Hi, I’m Michael Ryan and I’m an Incoming PhD student studying Artificial Intelligence at Stanford University. I’m fortunate to be doing NLP research as a member of Dr. Diyi Yang’s SALT Lab! I’m also a core contributor to StanfordNLP/DSPy – the library for programming not prompting LLMs. I’m on the optimizer team for DSPy and I am the co-creator of the DSPy MIPROv2 optimizer.

My research interest is Human-Centered NLP through two directions: LLM personalization for various cultures, languages, and individuals [1] [2] [3]. And leveraging humans for system design and feedback to make better AI systems. [4] Previously I was an undergraduate researcher in Dr. Wei Xu’s NLP X Lab at Georgia Tech and a research intern at Snowflake.

Have a look at my CV, or if you’re in a hurry, check out my resume!

Interests
  • LLM Personalization
  • Human-Centered NLP
  • Compound LM Systems
  • LM System Optimization
Education
  • PhD in Computer Science, 202X

    Stanford University

  • MS in Computer Science, 2025

    Stanford University

  • BSc in Computer Science (Intelligence & Systems/Architecture), 2023

    Georgia Institute of Technology

Research

GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning
SynthesizeMe! Inducing Persona-Guided Prompts for Personalized Reward Models in LLMs
Mind the Gap: Static and Interactive Evaluations of Large Audio Models
AudioJudge: Understanding What Works in Large Audio Model Based Speech Evaluation
EnronQA: Towards Personalized RAG over Private Documents

Teaching

Work Experience

 
 
 
 
 
Microsoft Corporation
Software Engineer Intern
May 2022 – August 2022 Seattle, WA
  • Designed and programmed static analysis tool in C++ for identifying security vulnerabilities throughout Windows OS.
 
 
 
 
 
Microsoft Corporation
Software Engineer Intern
May 2021 – August 2021 Virtual
  • Refactored existing codebase from .NET Framework to .NET Core.
  • Ported server-specific architecture to serverless functional units using Azure Durable Functions.
 
 
 
 
 
Uber Inc.
Software Engineer Intern
May 2020 – August 2020 Virtual
  • Implemented end-to-end testing in GoLang for bike, scooter, and moped rentals by building a simulated 3rd party CRUD API.

Projects

*
Knock Knock: Neural Joke Generation and Classification
A joke generator built on GPT-2 and joke classifier built on BERT for CS4650 Natural Language Understanding.
Knock Knock: Neural Joke Generation and Classification
Sign Assist
Sign Assist is a digital American Sign Language (ASL) interpreter. Using a Kinect camera and computer vision Sign Assist tracks user movements and patterns to detect different signs. Sign Assist can output these signs as both text and speech.
Sign Assist

Awards

Honors Program Distinction in Research
Awarded for completing Honors Level Coursework and Approved Research as an undergraduate at Georgia Tech
See certificate
Outstanding Undergraduate TA for Interactive Computing
Awarded for contributions to teaching excellence at Georgia Tech School of Interactive Computing

Contact