§ 01 — Field Notes, Rev. 2026.05

Will Macdonald

MedTech Advisor  ·  Chicago, IL  ·  AI on the Line

Notes from the seam between AI models and the workflows they get bolted onto. Mostly manufacturing and healthcare. Mostly what breaks in the second month, after the demo is over.


§ 02

Field notes.

LinkedIn — Recent posts

What's Actually Working with AI in MedTech Manufacturing—and What's Next

There's a gap between the AI conversation happening in conference keynotes and the AI conversation happening on the shop floor. I spend most of my time talking with medical device manufacturers—contract manufacturers scaling new facilities, OEMs automating production lines, companies trying to figure out how AI fits into an FDA-regulated environment. And the questions I hear most often aren't "should we adopt AI?" They're more practical: Where are companies actually seeing results? What's hype versus proven? And how do we move past pilot projects that never scale?

#MedTech #AI #Manufacturing
§ 03

Curated reads.

Source — AI Curation

Design and validation of a flexible framework for developing generic digital twins for Industry 4.0 | Scientific Reports

This article discusses the design and validation of a flexible framework for creating generic digital twins, a critical component of Industry 4.0. It highlights how digital twins bridge the gap between physical and digital spaces, driving advancements in manufacturing and the transition to Industry 5.0. The study emphasizes the transformative potential of digital twins in the ongoing physical-digital revolution.

Brave Search: manufacturing Industry 4.0 digital twin
§ 04

Notes in the margin.

Field Annotations — TIL

Better agent control using hierarchical agent.md/claude.md file structure:

#ai #agents
DEC 8 Source →

DSPy's BootstrapFewShot optimizer can learn from user feedback to improve prompts automatically - no manual prompt engineering needed.

#ai #dspy #prompt-engineering
NOV 30
§ 05

Standing order.

About
Will Macdonald

MedTech advisor in Chicago. Most of my work sits at the seam between AI models and the workflows they were bolted onto — manufacturing operations, regulatory and compliance, supply chain, and the long tail of things that quietly break in the second month after a pilot ships.

The coding here is so I understand the boring half of generative AI well enough to advise on it. I work on agentic systems, eval harnesses, and the integration plumbing that decides whether a pilot survives past the demo. When the model is the easy part, I'm interested in the rest.

When I'm not coding or advising, I'm in the kitchen, in the workshop, or behind a camera lens.

willmacdonald
Rev. 2026.05  ·  Chicago, IL