Skip to content

For clinics, provider groups and health systems

Take the desk work off your clinicians

Tiersel pulls the chart, triages the results inbox, finds an in-network specialist with an opening and builds the prior authorization packet from what is already documented. A clinician or staff member approves every order, message, booking and submission.

Live demo on synthetic data · Built on Jev by TypeSafe AI

An animated example types requests as if they were spoken and shows the card Tiersel builds for each one.

For the COO, VP of Clinical Operations or CMIO at a clinic, provider group or health system, with revenue cycle leaders as a second reader

Your clinicians spend more of the day on the record than with patients

In a time and motion study of ambulatory physicians, 49 percent of the office day went to EHR and desk work and 27 percent to direct time with patients. Prior authorization adds about 13 hours a week of physician and staff time. Much of that work is finding information that is already in the chart and moving it into another form, another queue or another portal.

13 hoursPhysician and staff time spent each week on prior authorizations, about 40 requests per physician per week.American Medical Association, 2025 prior authorization physician survey, 202549%Share of the office day ambulatory physicians spent on EHR and desk work, against 27% in direct time with patients.Annals of Internal Medicine (Sinsky et al., time and motion study in 4 specialties), 201680.7%Appealed Medicare Advantage prior authorization denials that were overturned in 2024. Only 11.5% of denials were appealed.KFF, analysis of CMS Medicare Advantage data, 2026$57.23Average administrative cost per claim for hospitals fighting denials in 2023. About 70% of denials were eventually paid.Premier Inc., survey of 280 hospitals, 2025

4 agent workflows

Built for the work your team already does.

Each workflow starts from a sentence. The agent calls your systems, shows its evidence, and waits for a person before anything consequential happens. Every card below is the real, working interface.

Workflow 01 · Primary care physician or care manager

Pre-visit chart review

Patient chartHealthcare

  1. 01 · Trigger

    Clinician asks by voice

    Between patients or before a visit, the clinician names the patient.

  2. 02 · Tool call

    Chart and results

    Reads problems, medications, allergies, recent visits and result trends from the EHR and lab system.

  3. 03 · Tool call

    Fills and schedule

    Checks pharmacy fill history for gaps in adherence and lists upcoming appointments.

  4. 04 · Reasoning

    Care gap check

    Compares the chart with your screening rules and flags overdue items, such as an A1c or a mammogram, with the source for each.

  5. 05 · Human approval

    Clinician picks the gaps to close

    Selected orders go to the ordering clinician to co-sign. Nothing is ordered without that approval.

    Nothing happens until a person approves

  6. 06 · Outcome

    Orders queued, chart updated

    Approved orders and scheduling requests land in the normal queues, and the action is logged.

What changes

The summary is ready before the clinician walks in, and overdue screenings are queued for co-signature during the visit instead of after it.

EHRLab information systemPharmacy fill dataScheduling

Workflow 02 · Referral coordinator or front desk

Referral and specialist booking

Provider finderHealthcare

  1. 01 · Trigger

    Staff ask for a specialist

    Specialty, plan, location, language and time window, in one sentence.

  2. 02 · Tool call

    Directory and network

    Searches the provider directory and checks in-network status with the patient's plan.

  3. 03 · Tool call

    Open slots

    Pulls new-patient openings from scheduling templates and confirms the referral on file.

  4. 04 · Reasoning

    Rank the options

    Ranks by network status, new-patient availability, language and distance. It never invents ratings.

  5. 05 · Human approval

    Staff pick a slot and approve

    The booking is prepared for one patient, one clinician and one time, and waits for approval.

    Nothing happens until a person approves

  6. 06 · Outcome

    Visit booked

    The appointment is held, the patient gets a confirmation, and the booking is logged.

What changes

Coordinators book an in-network specialist from one request instead of working through the directory, the network list and each clinic's schedule separately.

Provider directoryPayer network dataSchedulingReferral management

Workflow 03 · Ordering clinician

Results inbox triage

Results inboxHealthcare

  1. 01 · Trigger

    Clinician opens the day

    Asks which results need attention.

  2. 02 · Tool call

    Results feed

    Reads new lab and imaging results with reference ranges and flags.

  3. 03 · Tool call

    Chart context

    Pulls prior values, problems and medications for each patient.

  4. 04 · Reasoning

    Triage and draft

    Sorts critical, abnormal and normal by your critical value policy, prepares a callback script for criticals and drafts a plain-language portal message for abnormal results.

  5. 05 · Human approval

    Clinician edits and approves

    The clinician edits the draft and approves the send, or approves the callback task. Critical values are never sent as a portal message.

    Nothing happens until a person approves

  6. 06 · Outcome

    Patient informed, result filed

    The message is sent to the portal, a copy is saved to the chart, and the result leaves the queue.

What changes

Critical results surface first, and routine abnormal results arrive with a drafted patient message the clinician only has to review.

Lab information systemRadiology resultsEHRPatient portal

Workflow 04 · Prior authorization specialist

Prior authorization packet

Prior authorizationHealthcare

  1. 01 · Trigger

    An order needs authorization

    Staff start the request for the ordered service.

  2. 02 · Tool call

    Eligibility and policy

    Confirms coverage and that authorization is required, then loads the payer's criteria for the service code.

  3. 03 · Tool call

    Chart evidence

    Searches notes, imaging and medication history for evidence that meets each criterion, citing the document and date.

  4. 04 · Reasoning

    Find the gaps

    Marks each criterion met or unmet and names what is missing, such as physical therapy notes, before anything is sent.

  5. 05 · Human approval

    Specialist approves the submission

    Staff request the missing records, choose to hold or submit, and approve the packet.

    Nothing happens until a person approves

  6. 06 · Outcome

    Submitted with a complete packet

    The request goes to the payer with the attachments, and every step is logged.

What changes

Packets go out with the evidence the payer asks for, and missing documentation is caught before submission instead of in a denial.

EHRPayer eligibilityPayer prior authorization portal or APIDocument management

How it plugs in

An agent on your systems, one component per workflow.

Requests come in by voice, text or system event. The router picks the workflow, the agent calls your systems, and the answer lands as a live card where your team works. Nothing is written back until a person approves it.

In

VoicePush to talk, any browser
Typed requestSame intent, same card
System eventWebhook, queue or schedule

Tiersel

Intent routerPicks 1 of 4 workflows as you speak
Pre-visit chart reviewPrimary care physician or care manager
Referral and specialist bookingReferral coordinator or front desk
Results inbox triageOrdering clinician
Prior authorization packetPrior authorization specialist

Your systems

Clinical recordEHR (for example Epic, Oracle Health or athenahealth), Problem lists and notes
ResultsLab information system, Radiology reports
Payer and revenue cycleEligibility checks, Prior authorization portals or APIs
Access and schedulingScheduling system, Provider directory
Patient communicationPatient portal, Secure messaging

Out

Live cardIn your app, sidebar or kiosk
Human approvalClinician picks the gaps to close
Write-back + auditEvery call and decision logged

Workflow blueprint · Primary care physician or care manager

Pre-visit chart review

Starts when
Clinician asks by voice. Between patients or before a visit, the clinician names the patient.
Reads
Chart and results. Reads problems, medications, allergies, recent visits and result trends from the EHR and lab system.
Fills and schedule. Checks pharmacy fill history for gaps in adherence and lists upcoming appointments.
Decides
Care gap check. Compares the chart with your screening rules and flags overdue items, such as an A1c or a mammogram, with the source for each.
Waits for
Clinician picks the gaps to close. Selected orders go to the ordering clinician to co-sign. Nothing is ordered without that approval.
Writes back
Orders queued, chart updated. Approved orders and scheduling requests land in the normal queues, and the action is logged.
EHRLab information systemPharmacy fill dataScheduling

This workflow is one component

export const patient = defineSkill({
  id: "patient", category: "care",
  // when this workflow, and not a neighbour, is meant
  criterion: "A clinician or care team member asks to pull up a named patient's chart: problems, medications, labs, allergies, care gaps, visits. Not the user's own sleep, steps or heart rate, and not a data chart or stock chart.",
  examples: ["pull up maria gonzalez's chart", "open the patient chart for omar haddad"],
  match: (text) => score, // intent score, runs on every word
  parse: (text, ctx) => record, // entities + tool calls into your systems
  reply: (record) => "…",
  Card: PatientChartCard, // the live UI, with its approval step
});

The id, criterion, examples and reply are read live from the skill that renders the card above. Adding a workflow for your team means adding one of these: its intent, the calls it makes, what it says, and the interface your people approve in.

Fits your stack, answers to your people

Reads your systems. Acts only with approval.

What the agent connects to

Clinical record

EHR (for example Epic, Oracle Health or athenahealth)Problem lists and notesOrder entry

Results

Lab information systemRadiology reportsHL7 or FHIR results feeds

Payer and revenue cycle

Eligibility checksPrior authorization portals or APIsClearinghouse

Access and scheduling

Scheduling systemProvider directoryReferral management

Patient communication

Patient portalSecure messagingNurse line task queue

Product names are examples of systems in this category. Each deployment is connected to your systems through their APIs as part of the build.

A person approves every consequential step

Orders, portal messages, bookings and payer submissions wait for a named clinician or staff member. Each approval or rejection is recorded with who and when.

Deployed inside your environment

In production the agent runs in your cloud tenant or data center, under your Business Associate Agreement and your security controls, so PHI stays where it already lives. This demo uses synthetic patients only.

Every step is logged with its source

Each system the agent reads, each criterion it checks and each draft it writes is logged with the record it came from, so clinical and compliance leaders can audit what happened.

Least-privilege access

The agent reads only what a workflow needs and writes only to draft and approval queues, scoped by role and logged per request.

The cost of doing it by hand

Put your own numbers in.

Move the sliders to match your operation. Nothing is sent anywhere unless you choose to email it.

Each year

2,720

staff hours back, about 1.3 full-time people

$87,040

of staff time, at your hourly cost

Email me a pilot plan for these numbers

Default volume is about 10 physicians at the AMA survey's 40 requests per physician per week. Minutes per request come from the same survey: 13 hours for about 40 requests is roughly 20 minutes each. Hourly cost is the BLS May 2025 mean wage for medical assistants, $22.17, grossed up for benefits, which are about 30 percent of private industry compensation (BLS ECEC, June 2026). The agent finds evidence and assembles the packet while your staff still review and submit, so we count 40 percent of the time as recoverable.

Hours = prior authorizations a month × 12 × minutes each × share, ÷ 60. A full-time person is 2,080 hours a year. Default hourly cost from U.S. Bureau of Labor Statistics, OEWS May 2025, medical assistants, 2026.

Questions buyers ask

Straight answers.

  • Not in a deployment. The agent runs inside your cloud tenant or data center under your BAA and your access controls, and it reads your systems in place. This public demo uses synthetic patients only.

  • HHS does not certify software for HIPAA, so no vendor can honestly claim that. We deploy under your BAA and inside your security controls, and your security team can review the architecture and the audit logs during a pilot.

  • It is built around kinds of systems: EHR, lab and radiology results, scheduling, payer eligibility and prior authorization, and the patient portal, through standard interfaces such as FHIR and HL7 where you have them. We scope the exact connections for your stack before any build. The demo on this page is not connected to any live system.

  • Every finding cites the note, result or policy it came from, and every consequential step waits for a person. When evidence is missing, the agent says so, as with the missing physical therapy notes in the prior authorization demo, instead of guessing.

  • No. It does the lookup, matching and drafting. Your staff and clinicians review, edit and approve. The calculator assumes it takes 40 percent of the time on each task, not the job.

  • The demo on this page is free and uses synthetic data only. For your environment we start with one workflow and agree on scope and price in writing before any build.

See it on your own workflow.

Watch the healthcare workflows run end to end, or speak to the agent yourself. Then tell us which one to build on your data.

Scope a pilot with hello@sasid.ai
  1. 01

    Pick one workflow

    A short call to choose the workflow that costs you the most time, and who approves its actions.

  2. 02

    See it on your data

    We build that workflow against your sample or de-identified data so your team judges the real thing.

  3. 03

    Run it with people in charge

    Your staff use it on live work. The agent drafts and prepares; a person approves every consequential step.

  4. 04

    Decide from results

    You keep what measurably saves time and drop what does not.

Other operations

The same agent, other teams.