Impact

51

Student system tables unified into one model

4

Update paths, with AI used only where needed

2

Plain-language entry points for advisors and leadership

Turning stale degree audits into a live degree progress agent

Challenge

Advisors and the registrar at a large public research university rely on degree audits to show where each student stands. Each audit captures a single point in time.

Once a student adds a course, posts a grade, or changes majors, the audit falls out of date until it’s rerun. The data behind each verdict sits across more than 50 student information system tables, so seeing where any one student, or the full undergraduate population, actually stood took real effort.

The university wanted an always-current, governed view of degree progress that staff could query in plain language, built entirely inside Snowflake with Cortex AI.

Solution

Over two phases, OneSix built a Snowflake-native progress-to-degree pipeline. It models 51 source tables into a star schema, anchors each student to their most recent official audit, and detects every change since.

Each student then follows one of four update paths. If nothing has changed, the audit carries over as is. Simple changes are patched with SQL. Ambiguous changes, like a course that could fill two requirements, go to Cortex AI. Students with a new major or no audit at all get an on-demand reconstruction, flagged for re-audit.

AI has one narrow job: place each course into the requirement it fits best, or mark it unplaced. It never decides whether a requirement is complete, and every result is tagged with the path that produced it.

The result is requirement-level progress for every active undergraduate. Advisors use a Degree Progress Agent to ask which requirements a student still needs and which courses would fill them. Leadership uses Cortex Analyst to ask the same questions across the full population. The pipeline is documented and ready to schedule, with a runbook and roadmap for production.