Comify

Comify takes at least 3 pictures (and up to 9) and turns it into a comic strip. The app was built originally in ios/swift using AWS image rekognition. Once given the array of objects from AWS, we then selected a protagonist, setting, and adventures from the pictures to create a fantastical adventure for our user.

We then also addded a comic effect to the pictures they uploaded to make it more comic-booky, more likely to go with the story. The natural language processor for story construction uses a corpus of over a million words (1,167,335) to build its n-gram model and predict the next word in a story.

[Submission for MadHacks 2018]

Derm.AI

Derm.AI allows a user to take a picture of a mole and then compare that picture with our Machine Learning Model to predict if that mole is cancerous.

It originally used an image classification CNN model created on customvision.ai with the ISIC dataset and accuracy of 82%. We have now moved the model to AWS Sagemaker to improve accuracy. It also helps find a nearby dermatologist and oncologist professional and provide their contact information to schedule an appointment, using the Better Doctors API.

[Submission for Hackharvard 2018]

Blockchain Attendance

This is a decentralised attendance sheet built using Ethereum smart contracts with solidity and web3js, and integrated with ReactJS

[GitHub Demo]

Book Review

A Ruby on Rails Web project built with PostgreSQL and VueJS for curating through books and posting reviews that others can see, and reading the reviews of others.