Abhinay Karthik
All work

Case 3 of 4, Live

AyurAhaar

AI nutrition built on Ayurvedic principles: personalised diet plans that a doctor approves before a patient ever sees them.

Three interlocking rings: the three doshas a plan has to balance.

Overview

AyurAhaar turns traditional Ayurvedic nutrition knowledge into personalised, doctor-validated diet plans. It models the relationships between foods, doshas and diseases as a graph, generates recommendations from that structure, and delivers them through a mobile app for patients and a web portal for doctors.

React / React Native / Node.js / Express / MongoDB / Graph ML / Python

Problems, and what I built

  1. 01

    Encoding centuries of Ayurvedic knowledge as a structured, queryable graph

    A dosha, food and disease knowledge graph with embedding-based similarity search

  2. 02

    Balancing ML personalisation with clinical safety

    A dual-approval model: the AI drafts, a doctor reviews and edits before delivery

  3. 03

    Keeping mobile and web experiences at parity

    One shared component system across React Native and React

  4. 04

    Working with sparse, non-standard nutrition data

    A curation pipeline that turns Ayurvedic source texts into structured records

What it does

  • Meal plans generated from Ayurvedic dosha analysis
  • A doctor portal for approving plans and talking to patients
  • Telemedicine consultations by text and video
  • A dosha, food and disease knowledge graph
  • One patient dashboard on mobile and web

Result

Personalised, holistic nutrition guidance that scales, with the depth of a traditional system and the reach of software.