Interview Crew

The Open Source AI Career Coach for Engineers.

Interview Crew is a suite of free, open-source tools designed to help software engineers master technical interviews, build world-class resumes, and accelerate their careers.

We believe that high-quality career preparation shouldn't be paywalled. Whether you're a new grad or a Staff Engineer, you deserve the best tools to showcase your skills.

---

🚀 Features

🤖 AI Mock Interviewer

Practice makes perfect. Our AI conducts realistic technical interviews tailored to your level and role.

• Screening Rounds: Practice behavioral and rapid-fire technical questions.
• System Design: whiteboard complex architecture problems with AI feedback.
• Live Coding: Solve real world programming challenges with an AI pair programmer that hints, but doesn't solve.

📄 Smart Resume Builder

Stop sending resumes into the void.

• AI Analysis: Get instant feedback on your current resume based on industry standards.
• Tailoring: Automatically customize your resume keywords and summary for specific job descriptions.
• Templates: Clean, ATS-friendly templates designed for engineering roles.

🗺️ Career Roadmap

Don't practice randomly.

• Skill Gap Analysis: Find out exactly what you're missing for that Senior or Staff role.
• Curated Resources: Get a personalized learning path with the best articles, videos, and problems to solve.

---

💡 The Mission

Built by Engineers, For Engineers.

While our tools are free for candidates, our mission goes deeper. We are building an AI-First Recruitment Agency that flips the hiring model upside down.

Instead of recruiters spamming you, we use these tools to identify "A-Players"—engineers who demonstrate exceptional skill and growth. If you consistently perform well in our mock interviews, you can qualify for our Exclusive Talent Catalog.

• Get Scouted: We pitch you to top companies.
• Skip the Line: Fast-track past the resume screen directly to final rounds.
• Total Privacy: Your practice data is yours. You only enter the hiring pool if you explicitly opt-in.

---

🛠 Tech Stack

This project is built with a modern, type-safe stack focused on performance and developer experience.

• Monorepo: TurboRepo + pnpm
• Web App: Next.js 15 (App Router) - UI, Auth, DB, Webhooks
• Agent Service: Node.js / Express - Real-time AI Audio/Video processing
• Language: TypeScript
• Styling: Tailwind CSS v4
• Database: PostgreSQL (via Neon)
• ORM: Drizzle ORM
• Auth: Better Auth
• API: tRPC (Type-safe APIs)
• AI: OpenAI / Anthropic APIs (configurable)

📚 Documentation & Architecture

We maintain a log of all major architectural decisions in our Architecture Decision Records (ADRs).

• ADR-001: Voice & Video AI Stack Selection
• ADR-002: Monorepo Migration

---

⚡ Getting Started

Prerequisites

• Node.js (v20+)
• pnpm (v8+)
• Docker (for local services)
• Ngrok Account (for webhook tunnels)

Installation

1. Clone the repository:

   git clone https://github.com/interviewcrew/codebase.git
   cd interview

2. Install dependencies:

   pnpm install

3. Environment Setup: Copy the example environment file and configure your secrets:

   cp .env.example .env

   You will need keys for:

   • Database (Postgres connection)
   • Better Auth
   • OpenAI API
   • Stream.io (Video/Chat)
   • Inngest
   • Ngrok

Running Locally

We use a hybrid approach for local development:

• Docker: Runs the isolated Agent Service, Inngest, and Postgres.
• Host Machine: Runs the Web App (Next.js) for fast HMR.

1. Start Backend Services:

   docker compose up -d

   This starts:

   • agent: The AI Voice Agent (port 8000)
   • inngest: Event orchestrator (port 8288)
   • postgres: Local database (port 5432) - Optional if using cloud DB
   • ngrok: Tunnel for public access - Required for webhooks

2. Setup Database: Run migrations to the local or remote database:

   docker compose exec web pnpm db:migrate

3. Start Web App: The dev script automatically runs with the docker compose instructions & runs the app on http://localhost:3000. If any changes made on .env* files, you need to restart the web app using:

   docker compose restart web

   Open http://localhost:3000 to see the app.

4. Start Webhook Tunnel (Alternative): If you are not using the Dockerized Ngrok, run this manually to expose your local Next.js app:

   export NGROK_DOMAIN="your-static-domain.ngrok-free.app"
   pnpm --filter @interview/web dev:webhook

---

🤝 Contributing

We welcome contributions from the community! Whether it's fixing a bug in the mock interview logic or adding a new resume template, your help is appreciated.

1. Fork the repo.
2. Create a feature branch.
3. Commit your changes.
4. Open a Pull Request.

---

📄 License

This project is licensed under the MIT License.

---

Interview Crew — Code. Practice. Get Hired.