A Multimodal Job Search & Prep Agent

Inspiration

Nova Job Assistant is a multimodal, agentic AI system built using Amazon Nova models that helps job seekers streamline the entire job application process. Instead of manually reading job descriptions, rewriting resumes, and preparing for interviews separately, this AI assistant handles everything in one intelligent workflow.

What it does

Hours tailoring resumes Time analyzing job descriptions Struggling to identify skill gaps Preparing blindly for interviews

How we built it

We built Nova Job Assistant using Amazon Nova’s multimodal models with a React frontend and Node.js backend. It processes text, voice, and job screenshots, extracts structured data, matches resumes with jobs, generates tailored resumes and cover letters, and conducts mock interviews through an agentic workflow that automates the entire application process.

Challenges we ran into

Accurate Job Data Extraction Resume–Job Matching Logic Latency & Performance

Accomplishments that we're proud of

Built a True Agentic System Successfully Integrated Multimodal Inputs Accurate Resume–Job Matching Engine

What we learned

Agentic AI Design Handling Unstructured Data Prompt Engineering System Integration Challenges

What's next for A Multimodal Job Search & Prep Agent

Built With

• amazon-dynamodb
• amazon-web-services
• node.js
• nova
• pdf-parser
• react.js
• web-speech-api