LearnPath AI – Personalized Learning, Powered by AI

An AI-powered learning companion that identifies skill gaps, builds personalized learning roadmaps, and validates mastery to turn learning goals into real-world skills.

Comment

LearnPath AI – From Learning Goals to Personalized Mastery

💡 Inspiration

The internet has made education more accessible than ever. Platforms like YouTube, Coursera, and freeCodeCamp offer thousands of learning resources. Yet, one major challenge remains: learners often don't know what to learn, where to start, or what to learn next.

Two learners pursuing the same career goal may have completely different levels of knowledge. A beginner might need foundational concepts, while an experienced learner may only need to strengthen a few specific skills. However, most learning platforms recommend similar courses without fully understanding these differences.

We wanted to solve this problem by asking a simple question:

What if an AI system could understand what you already know, identify exactly what you're missing, and build a learning journey specifically for you?

That question inspired us to build LearnPath AI, an intelligent, multi-agent personalized learning platform that transforms learning goals into structured, measurable, and achievable learning journeys.

Our vision is to make personalized education accessible to everyone, regardless of their background, experience, or career aspirations.

What It Does

LearnPath AI goes beyond traditional course recommendation systems. Instead of simply suggesting a collection of learning resources, it guides learners through a complete personalized learning experience.

The platform follows a structured workflow:

Learner Profile → Goal Selection → Skill Assessment → Skill Gap Analysis → Personalized Roadmap → Learning Resources → Mastery Validation → Progress Tracking

1. Personalized Learning Goals

Learners begin by selecting their preferred domain, experience level, learning goal, and career objective.

The platform currently focuses on five technology domains:

  • Machine Learning
  • Data Science
  • Generative AI
  • Web Development
  • Cloud & DevOps

Users can create multiple learning paths under a single profile, allowing them to pursue different goals independently.

2. Domain-Specific Skill Assessment

Instead of relying entirely on self-reported experience, LearnPath AI evaluates the learner's actual knowledge through domain-specific diagnostic assessments.

These assessments examine important concepts and provide measurable evidence of the learner's current proficiency.

3. Intelligent Skill Gap Analysis

The Skill Gap Agent analyzes assessment performance to identify:

  • Existing strengths and competencies
  • Concepts requiring improvement
  • Missing prerequisite knowledge
  • Priority skills needed to achieve the target learning goal

This allows the system to focus on what the learner genuinely needs rather than repeating concepts they may already understand.

4. AI-Powered Learning Path Generation

Based on the learner's goals and identified skill gaps, the Learning Path Agent generates a structured roadmap divided into progressive learning phases.

Each phase includes learning objectives, relevant concepts, prerequisites, milestones, and practical activities.

The platform also explains why each phase is recommended and what the learner will gain by completing it, making recommendations transparent and understandable.

5. Curated Learning Resources

LearnPath AI connects learning phases with relevant educational resources, including free tutorials, documentation, and video-based learning materials.

The goal is to help learners discover useful content without spending hours searching across multiple platforms.

6. Mastery-Based Progression

Learning should be measured by understanding, not merely by completing content.

Our Mastery Agent evaluates a learner's understanding after a learning phase and supports progression based on demonstrated knowledge.

This encourages learners to build stronger foundations before advancing to more complex topics.

7. Personalized Progress Dashboard

The dashboard provides a centralized view of the learner's journey, including completed phases, remaining milestones, overall progress, and recommended next actions.

Each learning path maintains independent progress, making it easier to manage multiple learning goals.

How We Built It

We designed LearnPath AI using a modular, multi-agent architecture where specialized components handle different stages of the personalized learning journey.

Technology Stack

Frontend: React, JavaScript, and modern responsive UI components.

Backend: Python and FastAPI for RESTful APIs, application logic, and agent orchestration.

Database: SQLAlchemy-based data persistence for learner profiles, assessments, learning paths, and progress records.

AI Layer: AI/LLM-assisted recommendation and explanation capabilities combined with structured, rule-based evaluation.

Deployment: Vercel for the publicly accessible application.

Multi-Agent Architecture

Our system separates responsibilities across specialized agents:

Assessment Agent: Conducts domain-specific diagnostic assessments and evaluates learner performance.

Skill Gap Agent: Compares current proficiency with target competencies and identifies areas requiring improvement.

Learning Path Agent: Converts identified skill gaps into an ordered, goal-oriented learning roadmap.

Resource Recommendation Agent: Maps learning concepts to relevant educational resources and provides recommendation explanations.

Mastery Agent: Evaluates understanding after learning phases and supports mastery-based progression.

Progress Tracking Component: Aggregates learning activity and presents meaningful progress information through the dashboard.

End-to-End Workflow

  1. A learner signs in and selects a learning domain and career goal.
  2. The assessment engine evaluates domain-specific knowledge.
  3. The Skill Gap Agent identifies strengths, weaknesses, and learning priorities.
  4. The Learning Path Agent generates a personalized, phase-wise roadmap.
  5. Relevant educational resources are associated with each phase.
  6. The learner completes activities and takes mastery assessments.
  7. Progress is updated and visualized through the dashboard.

One important architectural decision was to use structured data exchange between agents rather than allowing every component to operate independently. This helps maintain consistency between assessment results, identified skill gaps, recommended learning phases, and progress records.

We also combined deterministic assessment scoring with AI-assisted personalization, rather than relying on a language model for every decision.

Challenges We Faced

1. Coordinating Multiple AI Agents

One of our biggest challenges was ensuring that different agents communicated consistently.

Assessment results needed to reach the Skill Gap Agent accurately, and the resulting skill-gap information needed to be correctly interpreted by the Learning Path Agent.

We addressed this by designing structured inputs and outputs and separating responsibilities between components.

2. Making Recommendations Truly Personalized

Generating a generic learning roadmap is relatively straightforward. Generating a roadmap that reflects a learner's existing knowledge, missing skills, and career goals is much more challenging.

We focused on connecting diagnostic assessment results directly to learning-path generation so recommendations could reflect individual learning needs.

3. Managing Multiple Learning Paths

A learner may pursue Machine Learning today and Generative AI tomorrow.

We needed to ensure that creating a new path did not overwrite existing paths and that assessments, phases, and progress remained associated with the correct learning goal.

This required careful handling of learner identities, domain-specific assessments, and learning-path identifiers.

4. Building Meaningful Progress Tracking

We encountered challenges in correctly representing phase completion, remaining modules, and overall progress.

We worked on making progress path-specific and ensuring that the dashboard reflected the learner's actual completion state rather than simply displaying static values.

5. Delivering a Complete Experience

Integrating onboarding, assessment, skill-gap analysis, roadmap generation, learning resources, and progress visualization into one coherent application was challenging.

We prioritized a smooth end-to-end workflow while keeping the architecture modular and extensible.

What We Learned

Building LearnPath AI taught us that effective AI applications are not just about generating intelligent responses. They require thoughtful system design, reliable data flow, and meaningful user experiences.

We gained practical experience in:

  • Designing modular, agent-oriented AI systems
  • Integrating AI capabilities with traditional backend services
  • Developing domain-specific assessment workflows
  • Translating assessment performance into actionable skill gaps
  • Creating explainable, goal-oriented recommendations
  • Managing multiple personalized learning journeys
  • Building responsive dashboards and user interfaces
  • Integrating frontend, backend, and data persistence into a deployed application

Most importantly, we learned that personalization becomes meaningful when recommendations are grounded in evidence about the learner, rather than assumptions alone.

Accomplishments We're Proud Of

We're proud of developing an end-to-end learning experience that connects skill assessment with personalized roadmap generation.

Rather than treating assessment, recommendation, and progress tracking as disconnected features, we designed them as parts of one continuous learning journey.

Our key accomplishments include:

  • Supporting five technology learning domains
  • Connecting diagnostic assessments to skill-gap identification
  • Generating structured, phase-wise learning roadmaps
  • Providing explanations for learning recommendations
  • Supporting mastery-oriented learning progression
  • Managing multiple learning paths for individual users
  • Presenting progress through an interactive dashboard
  • Deploying a publicly accessible web application

What's Next for LearnPath AI?

We envision LearnPath AI evolving into a more comprehensive intelligent learning companion.

Our future roadmap includes:

Adaptive Learning Paths: Dynamically adjust learning recommendations based on assessment results, mastery performance, and learning speed.

Expanded Domain Coverage: Support additional fields beyond technology, including business, design, finance, and other professional skills.

Conversational AI Tutor: Provide contextual explanations, answer learner questions, and assist with difficult concepts directly within each learning phase.

Advanced Resource Intelligence: Improve resource relevance using content analysis, learner feedback, and resource quality signals.

Project-Based Skill Verification: Evaluate practical projects alongside assessments to measure real-world application of knowledge.

Learning Analytics: Provide deeper insights into proficiency improvement, learning consistency, and estimated time to reach career goals.

Career-Oriented Recommendations: Connect acquired competencies with job-role requirements and industry skill expectations.

Our Vision

We believe education should adapt to the learner, not the other way around.

LearnPath AI aims to bridge the gap between where a learner is today and where they want to be tomorrow.

By combining intelligent assessment, skill-gap analysis, personalized recommendations, explainable learning phases, and mastery-based progression, we're working toward a future where every learner has a clear, structured, and personalized route to achieving their goals.

LearnPath AI — Don't just learn more. Learn what matters, in the right order.

Live Project

https://learnpath-ai-two.vercel.app/

Built With

Share this project:

Updates