GradePilot

plan your classes by semester, model weighted grade categories, and estimate your best‑case term GPA.

features

• semesters → classes → categories → items
• weighted grading (with optional “drop lowest” per category)
• drag to reorder items inside a category
• autosave (localStorage)
• optional syllabus ingest (Gemini) with a local fallback parser

run locally

npm install
npm run dev

then open http://localhost:3000.

optional: enable AI syllabus ingest (Gemini)

the app can call two API routes to extract grading categories from syllabus text / PDFs:

• POST /api/ingest-syllabus
• POST /api/ingest-syllabus-file

to enable that, create a .env.local:

GEMINI_API_KEY=your_key_here
# optional
GEMINI_MODEL=gemini-3-flash-preview

privacy note

if you use “autofill with syllabus”, your syllabus text/PDF is sent to Google’s Gemini API for parsing. if you don’t want that, you can keep using the app without a key and it will fall back to a local parser.

deploy

GradePilot is a Next.js app and deploys cleanly on Vercel.