Every appointment creates more work
A patient calls to book. Someone checks the schedule, prepares the visit, writes the note and assigns the codes. The clinician still has to review the record before the next patient arrives.
Sully.ai builds medical AI agents for that work before, during and after the visit. Its aim is more capacity for care, with clinicians and staff keeping the decisions that need their judgment. [1]
The agents work alongside the practice’s electronic health record, or EHR, so the encounter and the administrative work can stay together. [1]
A team around the visit
AI Receptionist handles patient calls and appointment work. AI Scribe turns the clinical conversation into a structured note for the provider to review. [1]
AI Medical Coder suggests codes from the encounter documentation and links them to the supporting note. The provider and billing team can review the code and its basis together. [1][2]
Sully also offers research and consultation tools, with its Consultant and Researcher marked as beta in the current product experience. Those tools support the clinician’s work rather than taking over clinical sign-off. [1]
EHR integrations carry the work into the systems the practice uses for patients, notes, appointments and codes. [1]
Nile Women’s Health Care: time back in the room
Dr. Hughan Frederick’s team at Nile Women’s Health Care was balancing charting with the attention patients needed during appointments. Providers kept turning from the conversation to the computer. [2]
Sully’s Scribe captures the visit and prepares the note alongside Athenahealth. Its Medical Coder generates suggested ICD-10 codes during charting, reducing a separate documentation step. [2]
The practice also uses Sully for repetitive patient-facing documents, such as work-excuse notes. Physicians, midwives and nurse practitioners work from a consistent documentation format. [2]
Nile reports 2.4 hours less charting per provider each day and an 18.5% increase in patients served after adopting Sully. Dr. Frederick has more attention available for the person in the room. [2]
Hillside: help new providers get started
At Hillside Primary Care, more than 40 providers worked across 18 Texas locations. New clinicians could spend months learning the practice’s documentation routines and EHR shortcuts. [3]
Sully’s Scribe, Interpreter and research tools gave them support from the start. Notes followed a consistent structure, and clinicians could work with multilingual conversations and look up supporting information during the visit. [3]
Hillside reports that onboarding fell from 12 weeks to two. The practice also reports up to four hours of daily administrative time saved per clinician. [3]
Those results come from Hillside’s use of the agents in its own practice. The work stayed centred on the clinicians and the records they needed to complete. [3]
Build the medical team around the patient
Ahmed Omar leads Sully.ai’s effort to give healthcare organisations a team of medical AI agents. The company is a Y Combinator S21 company. [4]
The product extends beyond a standalone note-taking tool. Calls, scheduling, documentation and coding become connected parts of the same practice operation. [1]
Practices can book a demo to explore the agents and the EHR integrations relevant to their team. Sully also provides an API for organisations building agents into their own systems. [1]
The direction stays close to the day’s work: take recurring administrative tasks off the care team’s list so the next patient receives their attention. [1][2]
Where Sully.ai ranks
Current rankings. Positions come from public evidence, and no company can pay for a place. How we rank
The bigger picture
Clinicians need complete records, and patients need to feel heard. Nile and Hillside show why those two goals belong together. [2][3]
Sully.ai is building the AI team around the visit, giving practices more capacity for care. [1]
Written by SOTA2 from Sully.ai’s public materials. Customer experiences and product plans are reported by Sully.ai. Ranking analysis and comparison data are from SOTA2.