SELECTED WORK/ResumeAI Engine
AI & Career Engineering·2026·Production Ready

ResumeAI Engine

AI-Powered Resume Optimization & Cloudflare Portfolio Publisher

ResumeAI solves the friction of manual resume customization. It provides a master profile management system, ingests resumes across PDF/DOCX/TXT/TEX and image formats via OCR, analyzes job postings to extract target ATS keywords, rewrites bullet points using quantifiable metrics, and publishes web portfolios to Cloudflare Pages with a single click.

ResumeAI Engine
5+ FormatsResume Parsing
<250msGroq Inference
CloudflareDeployment
NeomorphicUI System
The Problem Statement

Software job applicants spend dozens of hours rewriting resumes for different ATS engines, frequently getting rejected due to missing keywords, weak formatting, or lack of hosted web portfolios.

The Engineering Solution

A centralized platform where users maintain one master career profile, upload existing resumes in any format, run instant AI keyword tailoring against any job posting, and deploy custom web portfolios directly to Cloudflare Pages.

TECHNICAL ARCHITECTURE

Subsystems & Architectural Decisions

FRONTEND LAYER

React 18, Vite, TypeScript, Framer Motion, and custom Neomorphic dual-shadow tokens.

BACKEND LAYER

Node.js with Express 5 REST API handling parsing workers, AI prompting, and Cloudflare Pages API integrations.

DATABASE LAYER

MongoDB for master career profiles, work histories, and project records.

AIENGINE LAYER

Groq LLaMA models orchestrated via LangChain for ATS keyword correlation and action-verb enhancement.

Implemented Features & Capabilities

01

Multi-Format Document Ingestion (PDF, DOCX, TXT, TEX, and image OCR via Tesseract.js).

02

Job Description Tailoring: Instant keyword gap scoring and AI bullet refinement.

03

One-Click Cloudflare Pages Deployment: Automatically creates build artifacts and deploys live web portfolios.

04

GitHub Source Export: Pushes generated portfolio code directly to the user's GitHub repository.

05

Interactive Neomorphic Design System with responsive light and dark mode toggles.

Key Engineering Challenges Overcome

›Complex Document Parsing: Handled varied document structures by implementing multi-pass extractors with fallback regex parsing.
›Automated Cloudflare Deployments: Built a server-side packaging pipeline using JSZip and direct API upload streams.

Verified Technologies Used

React 18TypeScriptGroq AILangChainNode.jsExpress 5MongoDBFirebase AuthCloudflare Pages API