Inspiration
Job hunting in the modern tech ecosystem is more competitive than ever. Hundreds of applications are filtered out by Automated Tracking Systems (ATS) before they even reach human eyes. Many job seekers struggle with writing impactful summaries, phrasing bullet points with strong action verbs, or practicing for behavioral interviews. We wanted to create a solution that goes beyond traditional, static resume templates. Our goal was to build a comprehensive, AI-powered career co-pilot that doesn't just format resumes, but helps candidates optimize their content in real-time and practice with a personalized coach to land their dream job.
What it does
NextHire AI is an all-in-one career platform designed to make job preparation interactive and data-driven:
- Interactive Resume Builder: A sleek, premium dashboard where candidates can build and download beautiful, ATS-optimized resumes.
- AI Writer & Enhancer: Instantly transforms raw experience descriptions and summaries into professional, action-oriented, and ATS-friendly content with the click of a button.
- Interactive AI Career Coach: A real-time chat interface where candidates can ask for career advice, receive interview tips, and tailor their responses.
- Modern PDF Export: Instantly exports resumes into a clean, print-friendly PDF format. ## How we built it NextHire AI is built using a modern, scalable web stack:
- Frontend Framework: Next.js 16 (App Router) combined with React 19 and TypeScript for robust typing.
- Styling: Modern, custom Tailwind CSS (v4) with smooth glassmorphism, glowing accents, and dark-mode styling.
- Animations: Framer Motion for premium transitions, hover states, and micro-interactions.
- AI Core: Google Gemini 2.5 Flash integrated natively using the new
@google/genaiSDK for fast response times. - Deployment & CI/CD: Statically compiled via GitHub Actions and hosted on GitHub Pages for ultra-fast, zero-cost content delivery. We can mathematically model the resume alignment score $A$ against a specific job description's keyword vector $\mathbf{k}$ and the resume's text vector $\mathbf{r}$ using cosine similarity: $$A = \frac{\mathbf{r} \cdot \mathbf{k}}{|\mathbf{r}| |\mathbf{k}|} = \frac{\sum_{i=1}^{n} r_i k_i}{\sqrt{\sum_{i=1}^{n} r_i^2} \sqrt{\sum_{i=1}^{n} k_i^2}}$$ Our AI helper optimizes the resume text to maximize this alignment score $A \to 1$. ## Challenges we ran into
- Static Export & Environment Variables: Because we deployed the application on GitHub Pages as a static export (
output: "export"), we could not run server-side Next.js code or read runtime environment variables. We resolved this by configuring our GitHub Actions workflow to safely inject theNEXT_PUBLIC_GEMINI_API_KEYduring the automated build stage, baking it into the static client bundles. - PDF Layout & Formatting: Rendering dynamic, customized resume components into a clean, pixel-perfect PDF document can be challenging across different browser sizes. We optimized the canvas scaling and print styles to ensure that text lines remain crisp, spacing remains consistent, and page margins are preserved. ## Accomplishments that we're proud of
- Premium User Experience: Designed a beautiful, premium dark interface with rich aesthetics that stand out from boring, plain template platforms.
- Real-time AI Actions: Implemented lightning-fast text processing using Gemini 2.5 Flash, bringing instant feedback directly into the form editor.
- Secure Workflow: Successfully configured a fully automated build-and-deploy pipeline using GitHub Actions to inject variables and keep code cleanly decoupled. ## What we learned
- Prompt Engineering for ATS: We learned how to formulate highly detailed prompts for the Gemini model to ensure it outputs strictly structured text (like bullet points starting with strong action verbs) instead of generic conversational dialogue.
- Next.js Static Deployment: Mastered the nuances of compiling Next.js App Router applications to static exports, including handling paths, base assets, and environment variables on GitHub Pages. ## What's next for NextHire AI
- Mock Interview Simulator: Interactive mock interviews with scoring criteria based on responses.
- LinkedIn Profile Importer: Let users upload their LinkedIn profiles or PDF files to automatically populate their resumes.
- AI Mock Interview Coach: Add real-time audio chat capabilities using Gemini's multimodal features to practice speaking.
Built With
- framer-motion
- gemini-api
- github
- github-actions
- google-genai-sdk
- html2canvas
- jspdf
- next.js
- node.js
- react
- tailwindcss
- typescript

Log in or sign up for Devpost to join the conversation.