PathPilot AI

AI-powered academic planning assistant that helps college students choose their courses for the upcoming semester.

Students enter their academic background and the courses they are considering, and PathPilot AI generates a personalized course-selection report with a recommended schedule, risk analysis, and concrete next steps.

Live demo: https://pathpilot-ai-one.vercel.app/

---

Overview

PathPilot AI takes a few simple form inputs — school, major, year, GPA, completed courses, candidate courses, and a main goal — and turns them into a structured, readable report.

When an AI provider is configured, it generates a tailored report through the DeepSeek API on the server. When it is not, it falls back to deterministic rule-based logic, so the app always works.

Problem

Course registration is stressful and easy to get wrong. Students have to balance major requirements, GPA strategy, workload, prerequisites, and longer-term goals like transferring or applying to graduate school — usually with limited advisor time and scattered information.

It is hard to know whether a proposed schedule is realistic or risky before committing to it.

Solution

PathPilot AI converts those inputs into a clear course report. It recommends which candidate courses to take, flags workload and GPA risk, explains why the schedule makes sense, warns about risky course combinations, and suggests next steps — all in one place, with a guaranteed rule-based fallback so it is always usable.

Features

• Personalized course-selection report from a single form
• Recommended schedule drawn only from the student's candidate courses
• Risk analysis: overall risk, workload risk, and GPA risk
• Plain-language "Why This Schedule Works" explanation
• Course-by-course analysis and "courses to avoid taking together" warnings
• Actionable next steps and an academic disclaimer
• Graceful degradation: a rule-based report guarantees the app works without any API key
• Clean, responsive dark UI

Tech Stack

• Next.js — App Router, server components (Next.js 16)
• React — UI rendering (React 19)
• TypeScript — type-safe application code
• Tailwind CSS — styling (Tailwind CSS 4)
• DeepSeek API — AI-generated reports via server-side fetch (no SDK dependency)
• Vercel — hosting and deployment

How It Works

1. The user fills out the form at /course-report (school, major, year, GPA, completed courses, candidate courses, goal, and optional notes).
2. The form submits via GET to /report-result, passing the inputs as query parameters.
3. /report-result is a server component. It parses the inputs and asks the report generator for a structured report.
4. The page renders the report into clearly separated sections: Student Profile, Your Courses, Overall Recommendation, Course-by-Course Analysis, Courses to Avoid Taking Together, and Next Steps.

AI Integration and Fallback Logic

The report generator follows a single, predictable path:

1. AI path — If a DeepSeek API key is configured, the server calls the DeepSeek Chat Completions API and requests a JSON-structured report. The response is parsed and validated against the expected schema.
2. Rule-based fallback — The app silently uses the built-in rule-based report whenever:
   • DEEPSEEK_API_KEY is missing,
   • the API call fails or returns a non-success status,
   • the request times out, or
   • the response is malformed or fails validation.

This guarantees the demo is always functional, even with no key configured. The integration uses native fetch with no new npm dependency.

Security Notes

• The DeepSeek API call runs only on the server (inside a server component / server-only module).
• The API key is read from the DEEPSEEK_API_KEY environment variable and is never sent to or exposed in the browser.
• .env.local holds your real key locally and is git-ignored, so it is never committed.
• .env.example documents the variable name without containing a real key.

Local Development

# 1. Install dependencies
npm install

# 2. (Optional) enable AI reports
cp .env.example .env.local
# then open .env.local and set DEEPSEEK_API_KEY=your_deepseek_api_key

# 3. Start the dev server
npm run dev

Open http://localhost:3000 in your browser.

If you add or change DEEPSEEK_API_KEY while the server is running, stop it with Ctrl+C and run npm run dev again so the new value is picked up.

Other scripts:

npm run lint    # run ESLint
npm run build   # production build

Environment Variables

Variable	Required	Description
DEEPSEEK_API_KEY	No	DeepSeek API key for AI-generated reports. If unset, the app uses the rule-based report. Read on the server only — never exposed to the client.

Get a key at platform.deepseek.com.

Sample Input

You can try the demo with generic values like:

• School: Sample University
• Major: Computer Science
• Year: Sophomore
• GPA: 3.50
• Completed Courses: Intro to Programming, Data Structures, Calculus I
• Candidate Courses: Algorithms, Database Systems, Linear Algebra, Software Engineering
• Goal: Balanced schedule
• Notes: I want to choose a manageable schedule while making steady progress toward my degree.

Future Improvements

• Prerequisite checking against a real course catalog
• Save and compare multiple schedule options
• Accounts so students can revisit past reports
• Export the report as a shareable PDF
• Richer GPA-impact modeling based on course difficulty
• Support for additional AI providers

---

Disclaimer: PathPilot AI provides academic planning assistance only. Always verify requirements with your official school catalog and academic advisor.