Capstone Project
Industry capstone-scale work is now delivered with AI coding agents, so the graduating skill is no longer typing every line but specifying, decomposing, verifying, and taking accountability for machine-generated output. The capstone therefore keeps its role as the synthesis of the major while adding justified selection among candidate designs (including AI-driven ones), evidence of verification, and provenance for AI contributions.
Current catalog prerequisites — (CSC 250 or 2050).
Current description → proposed description
This course provides the student the opportunity to showcase computer science problem-solving skills by synthesizing an acceptable project. Students choose an acceptable problem and then fully implement the solution to that problem following professional programming practice. Students present their progress and project in both written reports and oral presentations.
Prerequisites: (CSC 250 or 2050).
This course provides the student the opportunity to showcase computer science problem-solving skills by scoping, implementing, and defending a substantial project. Students select an acceptable problem, weigh candidate solution approaches including AI-driven designs against requirements, cost, and risk, and justify the approach they adopt. Each project delivers a written specification, a milestone plan, and a working solution built under professional programming practice, including version control, testing, and documentation. Because contemporary software is increasingly produced with AI coding agents, students plan deliberately for machine-generated work: they decompose problems into well-specified tasks, define acceptance criteria and evaluation harnesses, verify code they did not personally write, and document the provenance of AI contributions so that authorship and accountability remain clear. Where agent components appear in the delivered system, students account for cost, latency, human-in-the-loop escalation, and prompt-injection or tool supply-chain risk. Students present progress and results in written reports and oral presentations, and reflect on the project as vocation undertaken in service of neighbor.
What changes
- Candidate solution approaches, including AI-driven designs, must be evaluated and justified
- Explicit plan for agent-generated work: task decomposition, acceptance criteria, evaluation harnesses
- Verification of code the student did not write becomes graded project evidence
- Provenance and attribution of AI contributions required for authorship and academic integrity
- Cost, latency, human-in-the-loop escalation, and prompt-injection/supply-chain risk accounted for where agents ship in the system
6 proposed outcomes, mapped to 10 program outcomes
Each outcome below is written to be observable and assessable, and each is mapped to the program learning outcomes for which it produces evidence.
Students will be able to evaluate candidate solution approaches to a self-selected capstone problem, including AI-driven designs, against stated requirements, cost, risk, and the Christian commitments the solution must uphold, in a written specification with defined scope and milestones.
PLO 4.2 requires students to propose and evaluate AI-driven solutions while upholding Christian values and Biblical truth, and this CLO instruments both halves: candidate AI-driven designs are evaluated and committed to in a written specification, and each is judged against the Christian commitments the delivered solution must uphold, which is the exit-assessed evidence the workbook records for this course.
Students will be able to implement a working solution under professional programming practice, including version control, automated testing, and documentation, decomposing the work into well-specified, independently verifiable tasks and, where AI coding agents are used, delegating them within stated cost, latency, and compute-resource budgets that the student justifies as responsible use.
Requiring the student to justify the cost, latency, and compute-resource budgets set for delegated agent work as responsible use is direct evidence for PLO 2.3, since the student must defend consumption of shared AI compute as ethical resource management rather than simply spend it.
Students will be able to verify machine-generated contributions for which the student remains accountable, using code review, acceptance tests, and evaluation harnesses that measure system behavior against the specification.
Building evaluation harnesses that measure system behavior against a specification, and explaining what those results show about the delivered system, is the skillful interpretation and explanation of AI output that PLO 6.1 requires.
Students will be able to critique the ethical and societal risks of the delivered system, including data privacy, bias, attribution of AI-assisted work, and prompt-injection or supply-chain exposure, against a Christian understanding of truthfulness and responsibility.
Critiquing the bias and societal exposure of the system's data-driven behavior is the critical analysis of data-driven outcomes and their societal implications that PLO 4.1 names; assessing that same system for fairness and moral impact is PLO 6.3; and judging data privacy and attribution of AI-assisted work against a Christian understanding of truthfulness is Christian values guiding professional ethical decisions under PLO 1.2.
Students will be able to articulate the project's rationale, design decisions, and results in written reports and an oral defense adapted to both technical reviewers and non-specialist stakeholders.
The written reports and oral defense are the written-and-spoken evidence PLO 5.1 requires, and adapting the same material for non-specialist stakeholders is the translation for broader audiences in PLO 5.3.
Students will be able to appraise the completed project as vocation, naming whom the work serves, what stewardship of AI tools required of the student, and what the student would do differently in professional practice.
Naming whom the work serves and what stewardship of AI tools demanded is the calling-and-stewardship evidence of PLO 1.3, and stating what the student would do differently is the structured self-reflection on one's own approach required by PLO 4.3.
Program outcomes this course reaches
Filled cells are program learning outcomes with at least one supporting course learning outcome in this course. Sparse coverage is expected — no single course carries all eighteen.
The workbook maps this course too
The proposal workbook's assessment map already assigns program learning outcomes to this course. The outcomes above were written to cover it.
| Program outcome | Workbook level | In this draft |
|---|---|---|
| PLO 4.2 Innovative AI Solutions | AE | covered |
I = Introduced · D = Developed · AE = Assessed at Exit