Career Counselor Multi-Agent Studio
Inspiration
Choosing a career is one of the biggest decisions students and early professionals face, yet many make that decision based on trends, social media, or incomplete information rather than their actual skills and interests. Most online career quizzes are static—they ask the same questions regardless of who you are and rarely explain why they recommend a particular path.
We wanted to build something that feels more like talking to a real career counselor. Instead of giving every user the same questionnaire, our system listens, adapts, challenges assumptions, and gathers evidence before making recommendations.
What it does
Career Counselor Multi-Agent Studio is an AI-powered career guidance platform that conducts an adaptive interview to understand a user's strengths, aspirations, and experience.
The platform can:
- Analyze resumes uploaded as PDF, DOCX, or TXT files.
- Generate personalized follow-up questions based on previous answers.
- Detect contradictions between career goals and actual experience.
- Explore alternative career paths when new evidence emerges.
- Automatically end the interview once it reaches sufficient confidence (up to 20 questions).
- Produce personalized career recommendations with a practical 14-day improvement roadmap.
Rather than relying on a single prompt, the application separates responsibilities across multiple AI agents that collaborate to build a more reliable recommendation.
How we built it
The project was built using:
- Next.js 15 App Router
- React
- TypeScript
- Tailwind CSS
- Groq and Google Gemini for AI reasoning and narrative generation
The system is organized into four specialized agents:
Interview Agent
Conducts a dynamic interview that adapts after every response instead of following a fixed questionnaire.
Resume Evidence Agent
Extracts meaningful skills, projects, technologies, and experience from uploaded resumes to provide objective evidence.
Reality Check Agent
Identifies inconsistencies between a user's stated goals and demonstrated experience, asking challenge questions when needed.
Career Match Agent
Combines interview responses and resume evidence to rank suitable career paths and generate a personalized 14-day action plan.
This modular architecture makes the reasoning process easier to extend and allows each agent to focus on a specific task.
Challenges we faced
One of the biggest challenges was making the interview genuinely adaptive.
Instead of pre-writing questions, every new question must consider:
- the user's dream career,
- resume evidence,
- previous answers,
- detected contradictions,
- current confidence level.
Designing prompts that produce consistent and relevant follow-up questions required significant experimentation.
Another challenge was balancing flexibility with interview length. We wanted conversations to feel natural without becoming unnecessarily long, so we introduced confidence-based stopping criteria with a maximum of 20 questions.
Resume parsing also presented challenges because users upload documents in multiple formats with varying structures. Extracting meaningful information consistently required careful preprocessing before sending the text to the AI agents.
What we learned
Building this project taught us that creating a good AI application involves much more than connecting an LLM to a frontend.
We learned how to:
- design collaborative multi-agent workflows,
- engineer prompts that maintain context across conversations,
- combine structured evidence with natural language reasoning,
- build adaptive interview systems instead of static questionnaires,
- create explainable AI recommendations rather than simple text generation.
Most importantly, we learned that separating responsibilities among specialized AI agents produces more transparent and maintainable systems than relying on one large prompt to do everything.
What's next
We see several exciting directions for the project:
- GitHub repository analysis for technical skill assessment.
- LinkedIn profile integration.
- Job market trend analysis.
- Personalized learning path recommendations.
- Internship and certification suggestions.
- Explainable confidence scoring and visual reasoning.
- Long-term career progress tracking.
Our goal is to evolve Career Counselor Multi-Agent Studio into an intelligent career companion that helps users make informed, evidence-based career decisions rather than relying on guesswork.
Log in or sign up for Devpost to join the conversation.