All projects

MoodMate

2025

Full-stack mental wellness app offering AI-driven insights based on journal entries.

PythonFastAPIMLflowPyTorchBERTGemini API

MoodMate is a full-stack mental wellness application that offers AI-driven insights and suggestions based on a user's journal entries, addressing the challenge of inaccessible therapy.

I designed the backend with FastAPI and MLflow, integrating a BERT text-classification model to extract emotion tags from entries, then used Google's Gemini API to generate personalized insights and feedback.

Highlights

  • Designed the backend with FastAPI and MLflow, integrating a BERT model for emotion extraction.
  • Leveraged Google's Gemini API for personalized insights and feedback.