Josh Mandel, MD

physician, programmer, and would-be pâtissier/poet. Chief Architect for #MicrosoftHealth and for @SMARTHealthIT.

Bsky:

[Josh Mandel, MD @JoshCMandel]

I am *constantly* typing some variation of "the output should be self contained and not make references to our conversation here -- the reader will not have this conversation in context"

[Ethan Mollick @emollick]

Its a consistent theory-of-mind failure in models that are otherwise suprisingly good at theory-of-mind

[Josh Mandel, MD @JoshCMandel]

With W3C Digital Credentials API, it's *finally possible* to design a health-data-sharing protocol that's idiomatic (standard FHIR-based resource sharing, questionnaire filling, etc) and works cross-platform (from web and mobile app requests to Android + iOS wallets)! See article for background + details on my updated "SMART Health Check-in Protocol".

[Josh Mandel, MD @JoshCMandel]

https://t.co/QJswd7s510

[Josh Mandel, MD @JoshCMandel]

https://t.co/QJswd7s510

[Josh Mandel, MD @JoshCMandel]

Yeah. Their app doesn't even support pinch to zoom.

[Keith W. Boone @motorcycle_guy]

I hate it that the @newyorktimes tries to upsell me every time I log in, and suggest their own app for reading. They are about to lose my subscription because of it.

If you are listening NY Times, you will lose a paying subscriber if you make me
Click six times just to get to my paid subscription content every log in.

[Josh Mandel, MD @JoshCMandel]

I wanted to share a small mental model that explains a *lot* of the gap between 1) "my data should be there in the API" and 2) "wait, where's my data?" And of course I can't resist making a few small recommendations :) See https://t.co/Jddq487vh0

[Josh Mandel, MD @JoshCMandel]

https://t.co/yT0kRKrzWD

[Josh Mandel, MD @JoshCMandel]

https://t.co/yT0kRKrzWD

[Josh Mandel, MD @JoshCMandel]

@maiab Do they mean sour?

[Josh Mandel, MD @JoshCMandel]

@thsottiaux If you mean Codex the GUI app... Linux support?

[Josh Mandel, MD @JoshCMandel]

https://t.co/kLEAZSZfXL

[Josh Mandel, MD @JoshCMandel]

There is *a lot* I want to know about where GPT-5.5 thought summaries come from. In the middle of debugging tool use outputs like:

> • Ran find /opt/android-sdk /usr/share/java -name '*guava*.jar' 2>/dev/null | head -20

I get the following thought trace (!!)

> • I need to look for a guava jar, so I'll start by searching online. I'm curious about the best places to find it. Should I check grocery stores, specialty shops, or maybe some online retailers? I wonder what brands are out there and if there's a specific type I should aim for, such as guava jam or preserves. I'll take a quick look and see what options come up!

[Josh Mandel, MD @JoshCMandel]

Hmm @AnthropicAI or knowledgeable users, does "full" here mean "including pre-resumption"?

>The design should have been simple: if a session has been idle for more than an hour, we could reduce users’ cost of resuming that session by clearing old thinking sections. Since the request would be a cache miss anyway, we could prune unnecessary messages from the request to reduce the number of uncached tokens sent to the API. We’d then resume sending full reasoning history.

[Josh Mandel, MD @JoshCMandel]

@dorofino I think you're talking about two very different layers of the stack. The shape and quality of the data that clients are authorized to see.

[Josh Mandel, MD @JoshCMandel]

How might SMART Permission Tickets help with patient self-access, proxy access, public health, and more? Overview / explainer / live demo here: https://t.co/AHJ8s5oBuA

[Josh Mandel, MD @JoshCMandel]

Airline platitude or undertaker platitude? "It is our honor to care for you on your journey."

[Josh Mandel, MD @JoshCMandel]

Even better than canned visualization app is adhoc report generation using the analysis JSON as input. https://t.co/YduYk5SZ1X is the generated visualization HTML, and https://t.co/W4a9nSN1iu is the Claude session that generated it (you can see it requires very little prompting).

@zakkohane

[Josh Mandel, MD @JoshCMandel]

This seemed like a fun task for Claude Opus 4.7! https://t.co/Z5qJwBNYwi has an example run on the claim "Hydroxychloroquine improves clinical outcomes in patients with COVID-19."

https://t.co/kn0aWevNVv describes the pipeline. I spent 3-4h on this with Claude Code managing 10 parallel copilot subagents for ~12h. It can run autonomously to analyze claims where the literature is findable in full-text open access sources.

https://t.co/vxIAWQPPYn is the underlying output data with nodes, edges, and judgments driving the UI.

[Josh Mandel, MD @JoshCMandel]

This seemed like a fun task for Claude Opus 4.7! https://t.co/Z5qJwBNYwi has an example run on the claim "Hydroxychloroquine improves clinical outcomes in patients with COVID-19."

https://t.co/kn0aWevNVv describes the pipeline. I spent 3-4h on this with Claude Code managing 10 parallel copilot subagents for ~12h. It can run autonomously to analyze claims where the literature is findable in full-text open access sources.

https://t.co/vxIAWQPPYn is the underlying output data with nodes, edges, and judgments driving the UI.

[Isaac Kohane @zakkohane]

I will personally cheerlead anyone who can bring AI-driven critical thinking to generate a citation network with these fine-grained _judgements_ in the typology links to any specified topic. There is a lot of AI work in this area but none to my knowledge could reproduce the figure below. Please prove me wrong. #citationNetworks

[Josh Mandel, MD @JoshCMandel]

@ick_real Mildred!