AI Career Coach

• 

  Start Azure function app locally

• 

  Generate technical interview question based on resume/job title

• 

  Evaluate an answer to an interview question based on relevance, tone, and quality

AI Career Coach

Inspiration

Preparing for interviews is stressful; most people don’t get enough feedback or realistic practice. I wanted to build something that simulates a real technical interview coach fast, personal, and always available.

What it does

AI Career Coach takes a resume or job title and generates custom mock interview questions using Azure OpenAI. It also evaluates candidate answers, rating them for relevance and tone and offering suggestions for improvement. It’s like having a career coach in your pocket.

How we built it

We used:

• Azure Functions for the backend API
• Azure OpenAI to generate questions and evaluate answers
• Node.js with an Express-like structure for logic and routing
• GitHub Copilot to scaffold function code, structure prompts, and assist with regex extraction
• Postman to test the API endpoints
• Markdown/README docs to package it all up for submission

Challenges we ran into

• Getting the right prompt engineering to guide OpenAI into producing structured and consistent feedback
• Parsing raw OpenAI output into clean JSON
• Managing Node.js version compatibility with Azure Functions runtime
• Environment variable setup for secure, repeatable local development

Accomplishments that we're proud of

• Two endpoints fully working
• Real-time AI-powered feedback with structured insights
• Full project setup, testing, and deployment-ready
• Clean API interface and a polished developer experience

What I learned

• Prompt engineering is critical to getting consistent responses from LLMs
• GitHub Copilot is an incredible force multiplier for prototyping under pressure
• Azure Functions + OpenAI is a powerful combo for serverless AI workflows
• Regex-based extraction from LLM output can be tricky but manageable with good formatting

What's next for AI Career Coach

• Add voice support using Azure Speech-to-Text and Text-to-Speech
• Save interview sessions and feedback to a database for progress tracking
• Add UI and mobile support for end-to-end experience
• Expand to behavioral and system design interview coaching

Built With

• azureai
• azurefunctions
• githubcopilot
• javascript
• node.js
• postman