S H A D H I N L A B

Loading

Book a meeting
Generative AI

DocAdvisor: a 24/7 AI health assistant for patient access

How we built an AI health assistant that understands symptoms in plain language, asks the right follow-up questions and points patients to nearby specialists.

All case studies
Illustrative product screen for DocAdvisor
Illustrative product screen
Client
DocAdvisor, our own product
Industry
Healthcare
Services
Generative AI
Platform
Web, docadvisor.xyz
Status
Live product
Summary

DocAdvisor is an AI health assistant built by Shadhin Lab. Patients describe symptoms in everyday language, answer a few follow-up questions, and get guidance on urgency, the right type of specialist and providers nearby. It supports doctors rather than replacing them.

  • Who: patients looking for timely guidance
  • Built: 24/7 symptom analysis and specialist matching
  • Stack: LangChain, RAG, Pinecone, ChromaDB, LLMs
  • Result: faster triage, projected
01 · The challenge

Patients wait too long to find the right care

Patients often wait weeks for appointments, struggle to explain their symptoms and get inconsistent direction, especially outside clinic hours.

Specialist shortages, complex referrals and distance make it even harder to reach the right doctor at the right time.

Before
  • Weeks of waiting for first guidance
  • No help outside clinic hours
  • Hard to know which specialist to see
After
  • Guidance in minutes, any time
  • Follow-up questions clarify symptoms
  • Nearby specialists suggested
02 · The solution

A first point of contact that leads to the right doctor

DocAdvisor reads symptoms written in plain language, asks follow-up questions about timing, severity and triggers, and builds a fuller picture before it recommends anything.

LangChain keeps context through the conversation, retrieval from curated medical sources keeps advice accurate, and AI agents route each case to a specialist category and nearby providers.

  1. 1Describe symptomsIn everyday words, no medical terms needed
  2. 2Answer follow-upsTiming, severity and what makes it better or worse
  3. 3Get guidanceUrgency and the right type of specialist
  4. 4Find a doctorSpecialists near you, with next steps
03 · Key features

What the platform does

24/7 health assistant

Guidance at any hour, without waiting for clinic opening times.

Real-time symptom analysis

AI interprets symptoms and spots patterns as soon as they are entered.

Intelligent follow-up questions

Targeted questions, similar to a clinical assessment, add context.

Location-based matching

Recommends relevant specialists and facilities in the patient’s area.

Personalised guidance

Recommendations reflect the patient’s own context and history.

Multi-language support

Patients use the language they are most comfortable with.

04 · Architecture

The technology behind it

Models
Large language models
Orchestration
LangChainAI agents and custom tools
Knowledge
RAG on curated medical sourcesPineconeChromaDB
Safety
EncryptionHuman review for high-risk cases
05 · Results

What changed

Projected figures

70–85% faster initial triage

Symptom assessment in under 5 minutes, compared with 20–30 minutes in conventional settings.

30–45% fewer non-essential visits

Patients understand severity before going to a clinic.

50–60% faster specialist discovery

Symptom analysis is linked directly to nearby specialists.

40–55% lower cost per triage

Thousands of concurrent users with little extra cost per user.

These are projected figures, based on industry benchmarks and the platform’s capabilities, as published by the DocAdvisor team. They are not measured results.

06 · Questions

Questions about this project

Does DocAdvisor replace doctors?

No. It is doctor-assistive. It gives early guidance and points patients to specialists, and complex or high-risk cases are escalated to qualified professionals.

How does it find the right specialist?

AI agents route symptoms to a specialist category, and custom location tools find providers and facilities near the patient.

How is patient data protected?

DocAdvisor collects only what it needs, encrypts data in storage and in transit, and does not share it with third parties without consent.

Get started

Want results like these for your business?

Tell us the problem. We will tell you how we would solve it, and what it would take to run it on AWS.

Not ready to talk? Email hello@shadhinlab.com and an engineer will reply.