Our company is interested in creating frameworks / templates for our pilot projects with social impact clients. Impact of this work is to maintain efficient operational execution in how we structure our client work.

We would like to collaborate with students to provide cohesive, appropriate details relative to the pilot project scaffolding from our Data Scientists. Students will write technical statistical documentation on methodologies and write code (likely in Python, using Jupyter and/or Marimo notebook) so we can achieve a fulsome pre-pilot understanding of the workflow involved.

This will involve several different steps for the students, including:

• Working with Data Scientist guidance to write code that assesses initial data type, structure, etc, especially on determining appropriate analytical tooling for client data use case categories
• Expanding on baseline Data Scientist framework document for synthetic data 'utility' metrics, i.e. how useful to a given statistical model the synthetic data is
• Completing all above in analytical objectives including but not limited to dimension reduction, relationship analysis, and clustering

By the end of the project, students should demonstrate:

• Understanding of key statistical modeling processes particularly as across different types and structures of social impact data
• Development of core applied math and computer science skills in statistics and programming AI/ML applications

Final deliverables should include:

• Source materials such as the data platform's back-end code (automated tooling by use case), if applicable, and accompanying technical documentation (data assessments, utility metrics) guiding our future pilot project work