Semester Autopilot
Your semester, on autopilot.
Drop in your syllabi. Get every deadline on one timeline, a grade-aware study plan, and a schedule that reroutes itself when life happens.
Free · no signup · local-first
Three minutes to a plan.
01
Drop your syllabi
PDF, pasted text, or a course-page link. The AI drafts every deadline, grading weight, and meeting time it can find — and flags what it can't.
02
Review every row
Nothing enters your semester unchecked. Low-confidence rows are highlighted with the exact source text, so a wrong date takes seconds to fix.
03
Live on autopilot
One timeline across courses, calendar export, and a study plan balanced by grade weight. Miss a day? It reroutes — and tells you what's at risk.
Features
The reroute
Tap “I'm busy” on any day — watch two weeks of work reflow around it, like GPS around traffic.
And when something genuinely no longer fits, it says so, in plain language — never silently overbooks you.
Grade what-if
“What do I need on the final for an A?”
Weight-aware projections from your actual syllabus. Drag a slider, watch your projected grade move. See exactly what skipping that quiz costs.
One timeline
Every course, every deadline, sized by how much it moves your grade. Exams read as diamonds; heavy weeks glow in the heatmap.
Calendar export
One click to .ics — lands cleanly in Google and Apple Calendar as all-day events, no timezone surprises.
Local-first
No account. Your courses live in your browser and export as JSON. The planner even works offline.
AI you can check. Math you can trust.
AI touches exactly one thing: reading your syllabus. It never invents a date — anything unclear is flagged needs a date with the verbatim source text, and you approve every row before it counts.
Everything else — the scheduler, grade math, calendar export — is deterministic, open-source code with 500+ tests, fuzzed across 100 random semesters and two timezones. Same inputs, same plan. Every time.
Free forever: no servers to pay for, runs against any OpenAI-compatible model — including a local one via Ollama, fully offline. MIT licensed.