Medical Director, Network Development Mount Sinai Hospital Network Chief, Cardiology Mount Sinai Hospital West
New York, NY
Joined December 2012
- Tweets41
- Following83
- Followers123
- Likes129
Today at #TCT2026 we built an app live on stage with the audience. They voted, wrote the features, an agent built it, and it was on their phones. Open source so you can run your own: github.com/MSWlabAI/vibe-cod… @TCTConference @crfheart @MountSinaiNYC
Today at #TCT2026 we built an app live on stage with the audience. They voted, wrote the features, an agent built it, and it was on their phones. Open source so you can run your own: @TCTConference @crfheart @MountSinaiNYC
Everyone wants an AI agent that can drive any app on your screen.
We're building one that can't, yet. On purpose.
Here's how we phase a screen-operating agent: guardrails first, autonomy last.
"Measured" means more than a pass rate. Our next eval gate includes negative tasks (prompt injection, ambiguity, forbidden actions, stale screens) that must pass 100%, plus do-nothing and random-click agents as baselines.
If a random agent scores well, your eval is broken.
Small local models are good workers and bad judges. A new paper: 0.91 on clean tasks, 0.00 when sources conflict. Our design: the model proposes, tested code decides and hands off to a human. arxiv.org/abs/2610.21270
Agents that operate clinical screens need guardrails first: a default-deny app allowlist, a cap on typed characters, a wall-clock limit, and no PHI in approval printouts. Read-only before write. That is how we are phasing our screen agent at LabQ.
From Prompts to Goals: Why AI Coding Agents Are Moving to /goal Workflows mswlab.ai/blog/goal-driven-a…
AI is finally cracking a 60-year-old problem in cardiology: pulling the heart’s magnetic signal out of hospital noise.
Great @WIRED piece on magnetocardiography — the work we’ve been doing at @MountSinaiNYC with @SandboxAQ.
Pivotal trial launching soon. Proud of this team.
wired.me/story/ai-is-learnin…
Jeffrey Bander MD retweeted
AI is transforming cardiovascular care, from earlier diagnosis to improved risk prediction and personalized treatment.
The CLIMB 2026 AI track builds practical understanding of how AI works, its limits, and real-world clinical use.
Apply: womenasonepulse.org/page/cli…
#CLIMB2026
ALT CLIMB SKILLS TRAINING PROGRAM Deepen your knowledge of Al in Cardiovascular Medicine Foundations of Al in cardiovascular medicine and clinical applications Use of Al in imaging, risk prediction, and clinical decision support Critical appraisal of Al tools, including strengths, limitations, and safety Apply now
The Great EHR Unbundling: Where Do Startups Win When Epic Absorbs Everything? mswlab.ai/blog/great-ehr-unb…
Jeffrey Bander MD retweeted
Don't miss Dr. @jeffBander at #ACC26 to learn about "Emergency Department Triage of Patients With Suspected Acute Coronary Syndrome Using Unshielded
Magnetocardiography."
#CardioX #Cardiology #Coronary @IcahnMountSinai @ACCinTouch @DLBHATTMD
The Great EHR Unbundling: Where Do Startups Win When Epic Absorbs Everything? mswlab.ai/blog/great-ehr-unb…
Modern AI Coding Architecture: Skills, Harness, Ralph mswlab.ai/blog/modern-ai-cod…
Jeffrey Bander MD retweeted
Be sure to attend Dr. @jeffBander's session on "Evolving the Scribe: Leveraging Agentic AI for Clinical Documentation" at #AHA25 on Monday, November 10.
#CardioX #CardioTwitter @American_Heart @IcahnMountSinai
