Skip to content

PRD.md

Latest commit

 

History

History
344 lines (224 loc) · 6.19 KB

PRD.md

File metadata and controls

344 lines (224 loc) · 6.19 KB

Product Requirements Document (PRD)

Project Title

AI‑Powered Personal Knowledge & Career Intelligence Platform

Author

Prabesh Khanal

Version

v1.0 (Student / Portfolio Build)

1. Purpose of the Product

The purpose of this product is to build a full‑stack, AI‑ready web application that allows users to store personal knowledge (notes, documents, projects, resumes), query that data intelligently, and receive AI‑driven career insights.

This project is designed to:

  • Master MongoDB and NoSQL data modeling
  • Use Mongoose effectively (schemas, validation, relationships)
  • Demonstrate AI system design aligned with AWS AI Practitioner concepts
  • Serve as a strong, explainable portfolio project

This is not intended to be a commercial SaaS. It is an interview‑grade engineering project.

2. Target Users

Primary User

  • Student / early‑career developer
  • Preparing for internships or entry‑level roles
  • Wants to organize learning and career material

Secondary User (Future Scope)

  • Any knowledge worker managing learning content
  • Users from any background (not only tech) — UI and defaults should avoid tech-only assumptions

3. Goals and Success Criteria

Goals

  • Allow flexible storage of structured and unstructured data
  • Provide AI‑assisted insights without paid services
  • Demonstrate clean backend architecture
  • Showcase NoSQL best practices

Success Criteria

  • User can store and retrieve all content reliably
  • AI pipeline works with mock or local inference
  • MongoDB schemas are defensible in interviews
  • Application is stable and demo‑ready

4. Functional Requirements (Step‑by‑Step)

Phase 1: User Management

FR‑1. User Registration

  • User can sign up with email and password
  • Passwords must be hashed

FR‑2. Authentication

  • JWT‑based authentication
  • Protected routes on backend

FR‑3. User Profile

  • Basic profile information
  • Skill interests (tags)

Phase 2: Knowledge Storage (Core NoSQL Learning)

FR‑4. Notes Module

  • Create, read, update, delete notes

  • Notes can be:

    • Plain text
    • Markdown‑like content
    • Tagged

FR‑5. Documents Module

  • Upload documents (PDF / text initially)
  • Store metadata separately from content

FR‑6. Projects Module

  • Store projects with:

    • Description
    • Skills used
    • Links
    • Status

FR‑7. Skills Module

  • Central skills collection
  • Reused across projects, resumes, and analysis

Phase 3: Resume & Career Data

FR‑8. Resume Storage

  • Store multiple resume versions

  • Structured sections:

    • Education
    • Experience
    • Skills

FR‑9. Resume Versioning

  • Maintain history
  • Allow comparison (basic)

Phase 4: AI Architecture (Free‑First Design)

FR‑10. AI Abstraction Layer

  • Central AI service interface

  • Supports:

    • Mock AI
    • Local inference
    • Cloud AI (future)

FR‑11. Text Processing Pipeline

  • Chunk large text
  • Store processed output

FR‑12. Embedding Placeholder System

  • Store vectors or simulated embeddings
  • Enable semantic‑style retrieval logic

Phase 5: AI Features

FR‑13. Knowledge Q&A

  • User can ask questions about stored content
  • System retrieves relevant data
  • AI generates grounded response

FR‑14. Skill Gap Analysis

  • Compare resume skills vs target role
  • Identify missing skills

FR‑15. Learning Recommendations

  • Suggest topics based on gaps

Phase 6: Analytics & History

FR‑16. AI Interaction History

  • Store prompts and responses
  • Timestamped

FR‑17. Progress Tracking

  • Skill growth over time (basic)

5. Non‑Functional Requirements

Performance

  • API response < 500ms for CRUD

Security

  • JWT authentication
  • Input validation via Mongoose

Scalability

  • Designed for future AI service upgrades

Cost

  • Must run at $0/month during development

6. Data Model Overview (MongoDB)

Core Collections

  • users
  • notes
  • documents
  • projects
  • skills
  • resumes
  • aiInteractions

Design Principles

  • Embed when data is owned by user
  • Reference shared entities (skills)
  • Avoid deep nesting

7. API Design (High Level)

Auth

  • POST /auth/register
  • POST /auth/login

Notes

  • GET /notes
  • POST /notes
  • PUT /notes/:id
  • DELETE /notes/:id

AI

  • POST /ai/query
  • POST /ai/analyze‑resume

8. Frontend Requirements (React + Tailwind)

Pages

  • Login / Register
  • Dashboard
  • Notes
  • Projects
  • Resume
  • AI Insights

UI Principles

  • Clean
  • Minimal
  • Data‑first

9. File Structure (Monorepo Style)

Root

/ai‑knowledge‑platform
│
├── backend
│   ├── src
│   │   ├── config
│   │   │   └── db.js
│   │   ├── models
│   │   │   ├── User.js
│   │   │   ├── Note.js
│   │   │   ├── Project.js
│   │   │   ├── Skill.js
│   │   │   ├── Resume.js
│   │   │   └── AIInteraction.js
│   │   ├── routes
│   │   │   ├── auth.routes.js
│   │   │   ├── notes.routes.js
│   │   │   ├── ai.routes.js
│   │   ├── services
│   │   │   ├── aiService.js
│   │   │   ├── embeddingService.js
│   │   ├── middleware
│   │   │   └── authMiddleware.js
│   │   ├── controllers
│   │   └── app.js
│   └── package.json
│
├── frontend
│   ├── src
│   │   ├── components
│   │   ├── pages
│   │   ├── hooks
│   │   ├── services
│   │   │   └── api.js
│   │   ├── context
│   │   └── App.jsx
│   └── package.json
│
└── README.md

10. Out of Scope (v1)

  • Payments
  • Multi‑tenant organizations
  • Real‑time collaboration
  • Heavy AI fine‑tuning

11. Future Enhancements

  • Cloud AI integration (Bedrock / OpenAI)
  • Advanced analytics
  • Interview simulator

12. Final Notes

This PRD intentionally prioritizes:

  • Architectural clarity
  • NoSQL correctness
  • AI‑readiness without cost

If built as specified, this project is sufficient for serious technical interviews.