COOCO is an AI-powered platform dedicated to reducing food waste and promoting sustainability by helping users manage their kitchen smarter. One key challenge we face is ensuring users make the most of their ingredients before they expire. Our goal for this project is to collaborate with interns to develop an AI-driven solution focused on optimizing ingredient tracking. Specifically, we want to leverage Artificial Intelligence (AI) and Machine Learning (ML) to predict which ingredients are at risk of expiring soon and suggest actionable ways to use them, such as personalized recipe recommendations.

This project offers learners a focused opportunity to apply AI/ML techniques to a real-world problem while contributing to COOCO’s mission of sustainability. Interns will work with our existing dataset—comprising user ingredient inventories and expiration dates—to build a predictive model that enhances our platform’s ability to minimize food waste.

The project involves the following steps:

• Exploring COOCO’s ingredient tracking dataset and its structure.
• Researching AI/ML techniques suitable for predicting expiration risks.
• Developing a simple AI/ML model to identify at-risk ingredients and suggest uses.
• Testing the model and refining it based on results.

By the end of the internship, interns should demonstrate:

• Understanding of the dataset: Ability to analyze and interpret COOCO’s ingredient inventory data, including quantities and expiration dates.
• Knowledge of AI/ML techniques: Familiarity with basic predictive modeling methods (e.g., regression, classification) relevant to expiration tracking.
• Application of AI/ML to a real-world problem: Successfully building a model that predicts at-risk ingredients and ties them to actionable suggestions.

Final deliverables will include:

• A concise report (2-3 pages) detailing:
• A brief overview of the dataset and its key features.
• The specific problem addressed (predicting ingredient expiration risks).
• The AI/ML methodology used (e.g., tools, algorithms).
• Results of the model, including accuracy or effectiveness.
• One practical recommendation for how COOCO can use the model moving forward.
• A working code file (e.g., Python script or Jupyter notebook) showing the model development and output.