PURPOSE: Outline course components

• Participants
• Supporting Materials
• Agreements
• Collaboration
• Grading
• Tentative Schedule
• Deliverable Details
• Promoting your Work
• Feedback
• Helpful Links
• Program Changelog

• Sponsor(s): This is the client. Provides project, relevant data, guidance.
• Faculty mentor(s): Assigned to each project, provides technical and professional guidance
• Students: Form and execute plan, complete the hands-on work, and provide deliverables to sponsor
• Capstone Director: Runs program end to end, provides oversight, can help with issues elevated above faculty mentor.
  Contact: Adam Tashman ([email protected])
• Impact Director: Partners with the Capstone Director to support and cultivate sponsor relationships, measure and promote the program.
  Contact: Rebecca Schmidt ([email protected])
• Capstone Manager: Assists the Capstone Director with communications, plans capstone events.
  Contact: Kylen Baskerville ([email protected])

• This course repo contains materials to support capstone projects, including:

  • Information on best practices
  • Templates for Agenda, Progress Reports, Final Paper, Presentation. Be sure to use these.
  • Grading Rubrics

• From Concepts to Code: Introduction to Data Science. Tashman, A.P. Chapman & Hall/CRC.
  Provides coverage of the topics that come up in capstone and more generally in the data science field. It is optional but recommended.

• Some projects have agreements that require student signatures (e.g., NDA, IP agreement, data agreement)

Meetings

• Reach out to your faculty mentor and sponsor for project kickoff.
  Meet the sponsor first, then the mentor.
• Schedule (ideally weekly) recurring meetings. Find times that work for everyone.
• If scheduling is hard some weeks (or during kickoff), go ahead and meet with a subset of attendees.
  Share meeting notes with the others.
• If you need to cancel a meeting, give at least 24hrs notice
• Once availability is shared, get meetings scheduled quickly
• You can use this template for meeting agendas

Sharing Information

• Figure out best modes of communication with sponsor and mentor.
  Example: As questions arise for the sponsor, you might have a shared doc that gets reviewed each week.
• Figure out the best place to store code and data.
  This should be discussed with sponsor and mentor. This is especially true when data is sensitive.

• Details in the Syllabus
• Students on same project may receive different grade based on participation, effort
• Faculty mentor does grading. Make your efforts clear to them:
  • In status reports, list your contributions
  • In final paper, list your contributions

• Fall 2025-Spring 2026 schedule. (Online MSDS).
• Spring 2026-Summer 2026 schedule . (Online & Residential MSDS).

1 | Faculty Mentor Briefing Report. [template]

• Summarizes activity and learnings between project kickoff and start of Capstone term
• Emailed to faculty mentor and Capstone Director during week 1 of Capstone term

2 | Progress Report I. [template]

3 | Progress Report II. [template]

• Reports spaced evenly over the term (due dates set by faculty mentor)
• Email to sponsor, mentor, Capstone Director

4 | Package containing all data products and research artifacts

• Shared with sponsors
• Students should coordinate with sponsors to determine best method for delivery
• Students to provide walkthrough, clear documentation so sponsor can use the work after project completion

5 | Final Report. [template] [rubric]

• Email to sponsor, mentor, Capstone Director
• Length: 6 pages maximum, excluding figures/tables/Appendices
• See template and rubric for details

6 | Presentation. [template] [rubric]

• Email to sponsor, mentor, Capstone Director
• Presented by all students at the Capstone Presentation event. Online MSDS event on Zoom, Residential MSDS event in person.
• See template and rubric for details

7 | Code Repository

• All code to be finalized and clearly documented
• Check sponsor preference for sharing location (e.g., GitHub private repo, shared drive)
• Students to ensure that sponsor can run the code for future use

General Guidelines

• Check with your sponsor about how the work can be promoted, as some projects may be sensitive
• Options can include listing on resume, posting on social media (e.g., LinkedIn), and publication

Publication

• Not a requirement but encouraged! Check with your sponsor first.
• Peer-reviewed journals may be preferable
• A non-peer reviewed option through the UVA Library is LibreOpen.
  Tagline: A robust scholarly repository that puts your UVA research center stage.
  Creates an immediate, persistent, worldwide-discoverable URL for your work (DOI – Digital Object Identifier).
• Another non-peer reviewed option is arXiv.
  arXiv is a curated research-sharing platform open to anyone.
  Students can create an account, prepare the manuscript in their acceptable format, and submit.

Authors

• Sponsors should be listed as authors if they wish to be included.
• Faculty mentors should be listed as authors if they wish to be included. If so, they are listed last.

Citing Work

• As a rule, all written or presented material must be your own unless you explicitly cite the work of others
• You may not copy or use the work of anyone else without proper citation

Voice and Style

• Do not use the first person. Use either second person plural (“we”) of passive voice where appropriate.
• Use direct and simple language.

Code

• Name all important code packages used

Figures and Tables

• Use block diagrams to describe data flow (pipeline) when appropriate
• Show mathematical formulae for core methods
• Tables should be legible and within the page’s columns
• Figures should be labeled clearly, crisp, and readable

Example Papers

• https://ieeexplore.ieee.org/document/11021089
• https://ieeexplore.ieee.org/document/7489317

• We want to hear from our capstone students, faculty mentors, and sponsors!
• The Program Director checks in with all parties throughout Capstone
• We send surveys at completion. Your thoughtful, timely feedback helps us improve Capstone.

• Technical documentation. Writing a good README for your software project.
• Dashboards. Streamlit: A faster way to build and share data apps.