Thread of side effects of “reasoning”

using Cline regularly has massively accelerated how Bitter Lesson-pilled I've become

This captures why I started building Osler-long before considering AI copilot features, I wanted to design clinical note software optimized for the ergonomics of structured thinking. Writing a great H&P medical note helps us generate and evaluate ideas, not just document them.

Just got a collection of o4-mini agents to collectively critique and work together on a piece of a research problem that I solved on my own that took about a week. It’s clear at this point that the latent capabilities of these models is much larger than is accessible via 1-shot

Some more reflections on using Osler in the clinical setting: it feels like a perfect balance of augmenting my clinical reasoning without losing control to AI. I originally built the tool without any AI functionality, more as a notepad to structure my clinical writing (and by

It’s amazing how much alpha there is in the Cline blog and posts by its engineers — so much value even applied to other domains besides coding agents

Building an app for capturing clinical pearls while on rounds/charting/studying. After you add a pearl, an AI agent will fetch relevant clinical guidelines and additional context. Any med friends interested in beta testing - let me know! Planning to launch (free) next week.