CareerNest – AI-Powered Career Coach

Transform your job search with AWS-powered AI that rewrites resumes, detects skill gaps and prepares you for interviews, all in one smart, scalable platform.

Comment

Inspiration

Every year, millions of students and professionals struggle with resumes, interviews and job applications that never get noticed. We’ve all been there, spending hours rewriting resumes and guessing what recruiters want. We wanted to solve this problem once and for all using AI and AWS cloud scalability.

That’s how CareerNest was born, a digital career coach that helps job seekers land their dream roles faster by combining AI, data and real-world career insights.

What it does

CareerNest is an AI-powered career optimization platform that helps users at every stage of their job search: 1) AI Career Coach (AWS Bedrock), Personalized, context-aware career conversations powered by Claude 3 Sonnet via AWS Bedrock. 2) Resume Optimizer - Upload your resume + job description → get instant improvement suggestions. 3) Behavioral & Technical Q&A Generator - Creates role-specific interview questions with answers. 4) Skill Gap Detection (AWS SageMaker) - ML pipeline predicts job match score, missing skills, salary estimate and time-to-job-ready. 5) ATS Score Analyzer - Tests how well your resume passes Applicant Tracking Systems. 6) Smart Project Recommender - Suggests mini-projects to strengthen weak areas in your resume.

Along with these an AI Mock Interview feature too. It’s like having LinkedIn Premium, a career mentor and ChatGPT, all in one app.

How we built it

We built CareerNest using a scalable AWS architecture combined with modern frontend tools: Backend: Node.js hosted on AWS EC2 Frontend: React + TypeScript (deployed on Firebase Hosting) AI Integration: AWS Bedrock (Claude 3 Sonnet) for generative AI ML Pipeline: AWS SageMaker for predictive skill and job match analytics Database & Auth: Firebase Authentication + Firestore Storage: Firebase Storage(for now considering the limited access) + optional S3 integration Payments: Razorpay API for subscription management Fallback AI: Gemini Flash for redundancy when Bedrock credentials expire All AWS credentials were securely managed through temporary session tokens provided for the hackathon.

Challenges we ran into

AWS credential management: We learned how to integrate temporary session tokens safely from the AWS Workshop Studio. Latency optimization: Bedrock requests initially took 8 seconds. We added async loading indicators and caching to improve UX. Prompt structuring: Claude requires strict role alternation, we had to implement prompt cleaning and conversation history truncation. Balancing models: We used a hybrid approach, AWS Bedrock for generative tasks, SageMaker for ML predictions and Gemini as fallback.

Accomplishments that we're proud of

1) Built a fully functional SaaS app in under 48 hours. 2) Integrated three AWS services (Bedrock, SageMaker, EC2) in one cohesive system. 3) Developed a conversational AI career coach with memory and context. 4) Achieved 99% uptime with intelligent fallback architecture. 5) Designed a scalable business model ready for launch.

What we learned

1) How to use AWS Bedrock for real-time generative AI tasks. 2) How to connect SageMaker endpoints for ML-based analytics. 3) How to design scalable, production-ready AI pipelines. 4) How to balance user experience, cost efficiency, and scalability.

What's next for CareerNest – AI-Powered Career Coach

1) Migrate file storage fully to AWS S3 for versioned resume management. 2) Add voice interview simulations using AWS Polly & Transcribe. 3) Deploy serverless components using AWS Lambda. 4) Partner with universities and career platforms for integration.

Share this project:

Updates