AI Interview Coach

Inspiration

I've seen too many people walk into interviews underprepared simply because realistic practice isn't accessible — so I built the tool I wish I'd had.

What it does

AI Interview Coach lets you pick a job role, answer 5 Claude-generated interview questions, and receive instant structured feedback on each answer, followed by an overall performance summary.

How I built it

I built it with a vanilla HTML/CSS/JS frontend talking to an Express.js backend, using the Anthropic SDK to call Claude for question generation, per-answer feedback, and session summaries.

Challenges I ran into

The trickiest part was maintaining coherent session context across stateless API requests — I had to carefully pass conversation history from the client with every call so Claude's feedback stayed consistent throughout the session.

Accomplishments that I'm proud of

I'm proud of shipping a fully spec-driven project end-to-end — from requirements and technical spec to tested, working code — in a single focused session.

What I learned

I learned how much a clear spec upfront accelerates development; having the scope and acceptance criteria defined before writing a single line of code eliminated almost all ambiguity.

What's next for AI Interview Coach

I'd love to add voice input so users can practice speaking their answers out loud, and eventually persist sessions so users can track their improvement over time.

Built With

• anthropic
• api
• claude
• css3
• express.js
• html5
• javascript
• node.js
• sdk