Michael Parker

Michael Parker

Founder, GLP-1 Observer

Associate Dean for Online Learning Research, Harvard Medical School (Former)
Assistant Professor of Medicine, Harvard Medical School (Former)

AI Applications

I'm building GLP-1 Observer, an independent tracker of the obesity-drug landscape. It includes a live clinical-trial dashboard and plain-English Field Guides. The GLP-1 Verdict provides a standardized assessment of efficacy, tolerability, and evidence strength. The GLP-1 Subway is a scroll-through explainer of how GLP-1 medicines work, set as a subway ride. Read more about why I'm building this.

Most recently, in my career at Harvard Medical School, I specialized in applying machine learning and large language models to analyze educational data, with a particular focus on understanding student learning patterns in online environments. My research combined quantitative analysis of structured learning data with AI-powered analysis of unstructured content like student comments, video transcripts, audio, images, and quiz question text.

I also founded and led HMX, an online learning product from Harvard Medical School. I built an interdisciplinary team, developed the product vision, and led the creation of an online learning platform and a range of unique, interactive online courses, from immunology to genetics to AI in medicine, with an emphasis on active learning and applying learning science to help people optimize their experience. I also have a strong technical background (originally computer science from MIT).

Contact: mparker [at] alum.mit.edu

Featured Research

AI-Enhanced Analysis of Feedback

Using large language models to analyze thousands of student feedback comments in online learning environments

Learn more →

What Don't You Understand? Identifying Student Misconceptions with LLMs

Using large language models with quiz performance data, student response patterns, and lecture transcripts to identify and characterize student misconceptions in online biomedical science courses

Learn more →

Interactive Data Visualization for Learning Analytics

Creating dynamic visualizations of educational data to find temporal patterns

Learn more →

Computer Control of Drug Delivery

Modeling drug diffusion with partial differential equations to enable computer control of drug delivery via intravenous infusion

Learn more →

Projects & Tools

GLP-1 Observer

Independent obesity-drug intelligence: live clinical-trial tracking, evidence-based drug verdicts, and plain-English field guides

View project →

Tools I Use

The development and productivity tools I use, including the coding agents and AI models

View project →

N-Gram Visualization

An application to help understand n-gram language models

Open project →

Recent Posts

Ride the GLP-1 Subway

September 30, 2026

I made the GLP-1 Subway, a scroll-through explainer of GLP-1s to help people explore how these drugs work. For people who are already familiar with GLP-1s, this resource can be shared to help explain them to others like patients, family members, or new hires. It’s harder to describe than to...

Working With AI Agents to Keep Fact-Dense Web Pages Fresh

September 16, 2026

How do you keep a fact-dense web page current when the facts change every week? Think of a Wikipedia article about something in the news. Wikipedia solves this with a swarm of editors. My pages need the same type of vigilance, and they have one editor: me.

Your Past AI Conversations May Be More Valuable Than You Think

March 30, 2026

Your past AI sessions are more valuable than you think. Over the last few months building a biotech tracking app and newsletter, I’ve accumulated hundreds of sessions in coding agents like Claude Code. I recently started treating those transcripts not as disposable history, but as a detailed record of how...