Knowtex Full-Visit AI Workflow

Diving deeper into

Caroline Zhang, CEO of Knowtex, on going from AI medical scribe to AI lab

Interview
doctors come to them saying, “I need this, and I have access to prove I’m getting a benefit.”
Analyzed 3 sources

This is what real bottom up pull looks like in healthcare, when a doctor can walk into procurement with usage data, faster chart completion, and extra billable work already visible in the EHR. Knowtex is not selling a vague AI promise. It is selling a tool doctors can use in live visits, then point to concrete outputs like drafted notes, staged orders, and more finished visits per day, which helps shorten a normally slow enterprise buying process.1 2 3

  • Knowtex is built around the full visit workflow, not just note writing. Before the visit it summarizes the chart, during the visit it drafts the note, and after the visit it stages orders, billing codes, follow up messages, and trial related tasks. That gives doctors more proof of benefit than a simple recorder app.1 3
  • The proof point matters because healthcare buying is usually split between user and buyer. In small practices a doctor can choose software alone, but in larger systems HIPAA agreements, IT review, and EHR integration push deals into enterprise sales. A doctor champion with clear productivity evidence can accelerate that handoff.2 3
  • This is also why the fight is shifting from best transcript to deepest workflow integration. Abridge won share by going deep with Epic and reached more than 60,000 clinicians across 100 plus health systems, while Knowtex is pitching cross EHR breadth and more actions around the visit. The winner is the product that removes the most clicks after the conversation ends.2 3

The next phase is a market where doctors expect free or bundled note drafting, and vendors win by proving they can automate the surrounding work inside the health system's actual software. That favors companies that can turn clinician enthusiasm into hard operational gains, then expand from documentation into coding, orders, and clinical decision support.2 3