AceMyInterview

AceMyInterview is a full stack interview practice application built with Next.js, TypeScript, Supabase, and Gemini AI.

Users can start a mock interview for a selected role, answer a set of questions, and receive structured AI feedback on a review page.

Features

• Role-based mock interviews
• Reusable global question bank
• Session-specific question selection (5 random questions per session)
• Interview history with score tracking
• Answer persistence
• AI-generated structured feedback (Gemini)
• Multi-dimensional scoring (communication, technical depth, problem solving)
• Strengths, weaknesses, improvements, and study topics
• Per-question tags, concept coverage, and reference answers
• Completeness signals for partial or brief submissions

Tech Stack

• Next.js (App Router)
• TypeScript
• React
• Supabase (PostgreSQL + Auth)
• Google Gemini API

Database Design

See SUPABASE_DB_SCHEMA.md for full details.

• sessions — one interview attempt per user
• questions — global question bank
• session_questions — links sessions to selected questions
• answers — user responses per question
• feedback — one row per session with overall_score, summary, and evaluation_json (v2)

Running Locally

Install dependencies:

npm install

Create a .env.local file:

NEXT_PUBLIC_SUPABASE_URL=your_project_url
NEXT_PUBLIC_SUPABASE_PUBLISHABLE_KEY=your_publishable_key
GOOGLE_GEMINI_API_KEY=your_gemini_api_key

If upgrading from an older feedback format, run the fresh-slate SQL in SUPABASE_DB_SCHEMA.md before testing.

Start the development server:

npm run dev

Open:

http://localhost:3000

Current Status

Implemented:

• Session creation and interview flow
• Question bank with join-table architecture
• Supabase auth and data persistence
• Gemini-powered feedback (v2 structured evaluation)
• Review page with scoring dimensions, SWI cards, and per-question breakdown
• History page with scores and same-role progress delta

Planned:

• PDF export of feedback
• Additional interview roles
• UI polish and role selector from database

Learning Goals

This project is used to learn:

• Next.js App Router
• TypeScript
• PostgreSQL data modeling
• Supabase
• Full stack application architecture
• AI integration workflows