Skip to content
View Eeshita-Pande's full-sized avatar

Block or report Eeshita-Pande

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Eeshita-Pande/README.md

Hi there - I am Eeshita 👋

🌱 I’m currently working on solving health and fintech problems - Building TheaAI, Product Monetization Lead at Cleo AI
🤖 I’m currently using applied AI in fintech and health, particularly building end-to-end applications for iOS (Swift)
🤔 Interested in health, longevity, and fintech. Passionate about optimising health — both actual and financial!

Languages

Python Swift

Data Science

Pandas NumPy Keras TensorFlow Scikit-learn Streamlit

Database

MySQL SQLite

Projects

TheaAI

Thea is a personalized wellness companion who offers actionable health recommendations visually, through chat, and on a home screen widget. We want to simplify everyday health decisions for people not obsessed with optimizing health outcomes.

This is an end to end project and I have been leading product, engineering, design, and marketing for TheaAI. I have built health models using Apple HealthKit to construct dynamic LLM prompts which are used to power the Swift app (Thea - the primary avatar, chat, and widget). I have implemented NLP / LLM concepts such as retrieval augmentation from Scratch in Swift, instead of using a framework like LangChain.

App Store Thea currently has 400 waitlist subscribers and 50 downloads

Demo

CleoAI Widget

I came up with the idea for a spending widget, and organised a small team to build it.

I built the widget in SwiftUI and worked with the other engineer to build a react-native bridge in order to fetch the data to populate the widget. We completed the full project (including widget design and implementation) in under a week.

Demo

Breast Cancer Prediction

I have trained several models to predict the likelihood of breast cancer when presented with industry specific inputs.

First, I explored the dataset in Google Colab, and then trained multiple models using the publicly available UCI Mammographic Mass Dataset

Next, I used Streamlit to create an interactive web-app which you can try out HERE

Demo

I used the following languages and frameworks:

Python

Streamlit

Pandas

NumPy

Scikit-learn

Keras

TensorFlow

Pinned Loading

  1. Projects Projects Public