PathFinder

Inspiration

As a student, I repeatedly noticed peers struggling with the same question: "Am I ready for this job?"

Beginners felt overwhelmed. Intermediate students felt unsure about missing skills. Advanced students didn't know if they were focusing on the right areas.

Most resume analyzers only give a score. Most learning platforms provide generic content. Very few tools offer clear, structured direction tailored to a student's current level.

I built PathFinder to solve this problem — not just to evaluate students, but to guide them step-by-step from where they are to where they want to be.

My goal was simple: Make career clarity accessible to students of all experience levels.

What it does

PathFinder is an AI-powered career guidance platform that:

  • Analyzes a student's resume (PDF upload)
  • Compares it to a selected job description
  • Identifies missing skills
  • Generates a structured 30-day personalized learning roadmap

Formally, if:

  • $S_r$ = Skills in the resume
  • $S_j$ = Skills required by the job

Then the skill gap is:

$$G = S_j - S_r$$

PathFinder does not stop at computing $G$. It transforms $G$ into an actionable roadmap with daily learning goals, recommended focus areas, and structured progression.

The system is designed to support:

  • Complete beginners
  • Intermediate learners
  • Final-year students preparing for internships
  • Students working individually or in teams

How I built it

PathFinder was built using a modern full-stack architecture:

  • Frontend: React / Next.js for a responsive user interface
  • Backend: Node.js for API handling and business logic
  • AI Integration: OpenAI API for skill extraction and roadmap generation
  • PDF Parsing: Resume text extraction from uploaded documents
  • Skill Matching Engine: Custom logic to compare resume skills with job requirements

The workflow:

  1. Resume is uploaded and parsed
  2. Text is processed and structured
  3. AI extracts relevant skills
  4. Job description is analyzed
  5. Missing skills are identified
  6. A personalized roadmap is generated

The system was designed modularly so new features such as collaborative mode, analytics, or adaptive learning levels can be integrated easily.

Challenges I ran into

1. Accurate Skill Extraction

Resumes are highly unstructured and vary in format. Extracting meaningful skills reliably required multiple iterations of prompt refinement and parsing improvements.

2. Avoiding Generic Roadmaps

Early AI outputs were too broad. I had to refine prompts to ensure roadmaps were:

  • Level-appropriate
  • Structured
  • Specific
  • Time-bound

3. Balancing Simplicity and Intelligence

The interface needed to feel simple for beginners while still offering depth for more advanced users.

4. Skill Matching Precision

Handling synonyms (e.g., "JS" vs "JavaScript") required improving the comparison logic to reduce false skill gaps.

Accomplishments that I'm proud of

  • Built a fully functional end-to-end system from resume upload to roadmap generation
  • Designed a solution that adapts to different student experience levels
  • Implemented both default and customizable roadmap options
  • Created a clean and intuitive interface accessible to beginners
  • Transformed abstract skill gaps into structured daily learning actions

Most importantly, I built a tool that genuinely helps students reduce uncertainty and gain clarity in their career preparation.

What I learned

Building PathFinder taught me:

  • The importance of precise prompt engineering in shaping AI outputs
  • How to design scalable full-stack systems
  • The complexity of working with unstructured real-world data
  • The impact of UX design on beginner accessibility
  • That technical solutions must remain human-centered

I also learned that clarity and usability are just as important as technical sophistication.

What's next for PathFinder

Next, I plan to expand PathFinder by:

  • Adding a collaborative team mode where students can build shared roadmaps
  • Integrating real learning resources (courses, documentation, tutorials)
  • Implementing progress tracking and milestone achievements
  • Adding adaptive difficulty scaling based on user performance
  • Supporting non-technical career paths
  • Building analytics to measure skill improvement over time

The long-term vision is to evolve PathFinder into a personalized AI mentor that grows alongside students throughout their academic and professional journey.

Built With

  • css
  • next.js
  • pdf.js
  • tailwind
Share this project:

Updates

Submission history