Softpiper Solutions

Healthcare · Customer conversations

Patient front desk and intake assistant

A non-diagnostic assistant for appointments, pre-visit forms, preparation instructions and reminders across web, phone and WhatsApp, on HIPAA-ready terms where needed.

  • ClaudeAnthropic
  • GeminiGoogle
  • Typical timeline: 6–10 weeks

The problem

Clinics lose hours of staff time to appointment calls, pre-visit forms, multilingual questions and reminder follow-ups.

  • Phones ring all day with scheduling and directions questions.
  • Patients arrive with intake forms incomplete.
  • No-shows leave gaps that a reminder could have prevented.

How we solve it

An assistant that schedules, completes intake, sends preparation instructions and reminders, and answers practical questions in the patient's language, with no diagnosis and a fast route to staff.

It plugs into EHR / HIS scheduling, WhatsApp Business, Google Workspace or Microsoft 365 and Phone line, so nothing you rely on today is replaced. The architecture below shows each step and which model does the work.

Architecture

From your stack, through the AI layer, to numbers you can track

Each step names the model or system doing the work. Writes to your systems wait for a person to approve them.
Patient front desk and intake assistant. Your systems (EHR / HIS scheduling, WhatsApp Business, Google Workspace or Microsoft 365 and Phone line) connect to Softpiper's AI layer, which runs these steps: Patients reach out (WhatsApp Cloud API · Web chat); Understand and route (Claude Sonnet); Schedule (EHR / HIS API); Intake and preparation (Strict tools); Voice and call notes (Gemini Live · Transcribe); Reminders (WhatsApp templates). Outcomes are measured as no-show rate, calls handled without staff, intake completed before the visit and patient wait time.

Your stack

  • EHR / HIS scheduling
  • WhatsApp Business
  • Google Workspace or Microsoft 365
  • Phone line

AI layer

Softpiper AI layer

  • Claude
  • Gemini
  1. Patients reach out, using WhatsApp Cloud API · Web chat

    Patients write on WhatsApp or the website, or call the voice line.

  2. Understand and route, using Claude Sonnet

    Requests are classified; anything urgent or clinical goes straight to staff with emergency guidance.

  3. Schedule, using EHR / HIS API

    Slots are read from, and booked into, your scheduling system through its API.

  4. Intake and preparation, using Strict tools

    Intake forms are completed in conversation and preparation instructions sent before the visit.

  5. Voice and call notes, using Gemini Live · Transcribe

    Calls are handled in the caller's language and transcribed with speaker labels.

  6. Reminders, using WhatsApp templates

    Approved WhatsApp templates send reminders and follow-ups.

Outcomes you measure

  • No-show rate
  • Calls handled without staff
  • Intake completed before the visit
  • Patient wait time

Why this model

Claude and Gemini, each where it leads

We choose per task on capability, cost and where your data must stay. Final versions are picked at kickoff.
ClaudeAnthropic

Claude handles scheduling, intake and FAQ conversations with strict tools. Where protected health information is involved we deploy on the first-party Claude API under a signed BAA, which supports HIPAA-ready use. Claude Platform on AWS, Microsoft Foundry, Claude Code and Managed Agents are not covered, so we keep PHI away from them.

GeminiGoogle

Gemini's Live API and transcription models handle multilingual voice calls and call notes, under your own Google Cloud agreements.

Delivery

What you get, how long it takes, and how we keep it safe

Guardrails

  • No diagnosis or treatment advice; clinical questions go to staff.
  • Emergency keywords trigger an immediate escalation message and a staff alert.
  • Only the fields the workflow needs are collected.
  • PHI is processed only on BAA-covered services and never placed in structured-output schemas.
  • Built to align with India's DPDP Act as its obligations phase in.

Plus the guardrails on every build: facts from your systems, approvals for writes and money, a human one step away, audit logs and evals. See all

Deliverables

  • Scheduling and intake assistant for web and WhatsApp
  • Optional multilingual voice line
  • Escalation rules and staff alerts
  • Reminder templates
  • Compliance documentation for your review

Typical timeline

  1. Build, including compliance review6–10 weeks

A planning range for a first production release. We confirm scope, price and dates in a tailored proposal.

KPIs to track

  • No-show rate
  • Calls handled without staff
  • Intake completed before the visit
  • Patient wait time
  • Staff time per appointment
  • Patient satisfaction

Goals we agree with you and measure from a baseline. They are what we track, not results we promise.

Related services

The services behind this playbook

More playbooks

Related playbooks

All playbooks

Pip · Softpiper's AI

Ask Pip about this playbook

Ask Pip how this would work for you

Pip is Softpiper's AI. Tell it what you run today and it will walk through how this playbook fits, what we would build first and a tailored price range. Or book a call with the team.

Book a call