MatchMate 💖

An android app that simulates a Matrimonial App by displaying match cards similar to Shaadi.com's format. Built with Kotlin, Jetpack Compose, Room, and Retrofit, users can scroll through profiles, accept or reject matches, and view previously accepted matches.

Demo

Screenshots and demo clips: https://drive.google.com/file/d/1X23BA5nY9RlgFN0ZT_YYEz8_k5BF7cAX/view?usp=sharing

Features

• Home Tab to display profiles in cards with details. In background it fetches user data like name, age, city(From API) and then displays them with additional details like education, religion, community and profession (As observed to be the some of the prime factors considered during match making)
• Matches Tab to view accepted/rejected profiles
• Profile Tab to display user profile (Dummy as of now)
• Local match score based on age/religion matching
• Offline support with Room database (holds 50 undecided profiles)
• Network check with graceful error handling and retry logic
• Light/dark mode toggle

API Reference

Get all items

  GET https://randomuser.me/api/?results=10

Note: After data is fetched from the above API, a match score is calculated, random religion, caste and profession is added (from a list of local list of religions/caste/profession)

Installation

1. Clone the repository and open in android studio
2. Let the gradle sync and build
3. Run the application on an emulator or physical device

Architecture and folder structure

This app uses MVVM architecture with a clean separation of concerns.

Reasons:

• separation of concerns models: data layer views: ui / presentation viewmodels: business logic
• states are managed with jetpack compose allowing a unidirectional flow of data
• survives configuration changes due to use of viewmodel
• easy testing as all the layers are separated out

Link to folder structure: https://limewire.com/d/AAUaF#8bRXtDF9Yw

Tech Stack

Jetpack Compose
Coil for image rendering (supports caching by default and light weight)
Retrofit for Network Calls
Material3 UI
Room DB (for offline local storage)

Libraries and versions

agp = "8.7.3"
kotlin = "2.1.10"
compose = "1.8.2"
material3 = "1.3.2"
activity = "1.10.1"
navigationCompose = "2.9.0"
coilCompose = "2.2.2"
coreKtx = "1.16.0"
appcompat = "1.7.0"
material = "1.12.0"
junit = "4.13.2"
junitVersion = "1.2.1"
espressoCore = "3.6.1"
room = "2.7.1"
retrofit = "2.9.0"
gsonConverter = "2.9.0"

To Do

• Add more fields for profiles like hometown and eating preferences
• Stimulate flaky network
• Add authentication module
• Cloud database syncing
• Swipe gesture for a more modern look
• Sort/filter profiles based on match score
• Profile editing

Match Score logic

Currently a simplified logic has been used. More advanced logic can be integrated easily.

Each user profile receives a match score based on:

Age similarity (60 score) + Religion match (40 score)