PrepForge is an AI-powered interview preparation platform designed to simulate real-world technical interviews through structured practice, feedback, and performance tracking.
PrepForge is built around the concept of a forge:
“Through repetition, pressure, and focused effort, strength is forged.”
Users:
- Enter the forge (start sessions)
- Train under pressure (timed interviews)
- Refine their skills (AI feedback + recommendations)
- Track growth over time (dashboard + analytics)
-
User receives AI-generated problem:
- Title
- Description
- Constraints
- Example test cases
-
User must:
- Restate the problem
- Explain approach
Rules:
-
Time limit: ~5 minutes
-
Auto-submit on timeout
-
AI evaluates:
- understanding
- approach validity
-
DOES NOT block progression
-
User enters coding environment (Monaco-style editor)
-
Features:
- Starter code (imports + class)
- Language selection
- Run example test cases
- Autosave
- Timer (20–30 minutes)
Outcomes:
- Early submission OR
- Auto-submit on timeout
AI evaluates:
- Correctness
- Code quality (clean vs spaghetti)
- Time complexity
- Space complexity
- Closeness to solution
Also:
- Compares Stage 1 vs Stage 2 (planning vs execution)
Outputs:
- strengths
- weaknesses
- recommendations
- next problem suggestions
- 4–5 questions per session
- 90–120 seconds per question
- Failure
- Teamwork / conflict
- Leadership
- Technical challenges
- Learning new tools quickly
- Voice (speech-to-text)
- Typed response
AI scores:
- STAR structure
- Clarity
- Specificity
- Relevance
- Time management
-
Per-question feedback
-
Final summary:
- average score
- strongest/weakest areas
- coaching insights
- recommended practice areas
Purpose:
“How am I doing?”
- Total sessions
- Avg coding score
- Avg behavioral score
- Current streak
- Score trend chart
- Weak areas
- Recent sessions
- Active recommendations
- Motivation card
Purpose:
“What have I done?”
- Type (Coding / Behavioral)
- Score
- Status (Submitted / Auto-submitted)
- Date
- Duration
- Summary
- Type
- Date range
- Score
- Completion status
Purpose:
“What should I improve?”
- Active recommendations (top)
- Past recommendations (grouped by session)
- Title
- Reason
- Priority
- Source session
- CTA (Start Practice)
-
Dashboard
-
Practice
- Coding Interview
- Behavioral Interview
-
Sessions
-
Recommendations
-
Profile
| Page | Purpose |
|---|---|
| Dashboard | Progress overview |
| Practice | Start sessions |
| Sessions | History |
| Recommendations | Improvement plan |
Dark, focused, intense — representing discipline and growth.
- Background: #0B0B0F
- Secondary: #12121A
- Accent: #FF6A00 (forge fire)
- Glow: #FFD166
- Text: #EAEAEA
-
Title: PREP FORGE
-
Quote:
“Consistency, Focus, and Drive forge greatness.” Enter the Forge.
-
CTA Button:
ENTER THE FORGE
| Standard | PrepForge |
|---|---|
| Start | Enter |
| Submit | Forge Result |
| Practice | Train |
| Sessions | Forged Sessions |
| Recommendations | Refinements |
| Stage | Name |
|---|---|
| 1 | Understanding the Metal |
| 2 | Shaping the Blade |
| 3 | Final Tempering |
- Spring Boot + Spring Security
- JWT Access Tokens
- Refresh Tokens (HttpOnly cookies)
- PostgreSQL
- email + password
- password hashed (BCrypt)
- email verification required
- returns access token
- sets refresh token cookie
| Token | Purpose |
|---|---|
| Access Token | API authentication |
| Refresh Token | Session persistence |
- POST /api/auth/register
- POST /api/auth/login
- POST /api/auth/refresh
- POST /api/auth/logout
- GET /api/users/me
- Password hashing (BCrypt)
- HttpOnly cookies
- Role-based access
- Generic login errors
- Rate limiting (future)
- Email verification
- Password reset email flow
- Auth
- Sessions
- User data
- REST endpoints
- Question generation
- Coding evaluation
- Behavioral evaluation
- Recommendations
- Score aggregation
- Trends
- Dashboard data
- Run test cases
- Validate correctness
- coding.question.generate.requested
- coding.submission.received
- coding.feedback.generated
- behavioral.answer.submitted
- behavioral.feedback.generated
- recommendation.generated
- Docker (containers)
- Docker Compose (local orchestration)
- Kafka (event streaming)
- Redis (caching + rate limiting)
- PostgreSQL (primary DB)
- GitHub Actions (CI/CD)
- Railway / AWS (deployment)
Dashboard → Practice → Session → Feedback → Dashboard
Dashboard → Recommendation → Practice → Improve
Sessions → Review → Practice Again
- System Design Interview Module
- Real-time code execution engine
- AI voice tone analysis
- OAuth (Google login)
- Multi-factor authentication
- Resume integration
- Peer mock interviews
PrepForge is not just a practice tool.
It is:
“A structured training environment where engineers improve through repetition, pressure, and intelligent feedback.”