🚀 Automating recruiter follow-ups with AI — because great candidates deserve to be remembered.
QuickFollowup is an AI-powered Chrome extension that helps job seekers automatically generate personalized follow-up emails, track every application, and find message-able recruiters on LinkedIn — all in one flow.
The system detects job submission events, extracts key details (company, position, and description), and uses LangGraph + OpenAI API to craft custom follow-ups.
It also leverages a Chrome background service worker to locate recruiters and determine if they’re DM-able.
Like many job seekers, I noticed that cold applications often disappear into a void.
Manually following up — switching between ChatGPT, LinkedIn, and spreadsheets — was exhausting and chaotic.
I built QuickFollowup to make this process automatic, human, and even fun:
- Auto-generate tailored emails instead of copying from ChatGPT.
- Auto-open recruiter profiles that can actually receive messages.
- Auto-track each application’s status — all while earning coins to care for your virtual Tamagotchi-style pet.
The goal is to make job follow-ups consistent, personalized, and stress-free.
QuickFollowup consists of four main components:
-
Frontend (Chrome Extension – React + TypeScript)
- Detects job application submissions and scrapes job details from LinkedIn, Workday, etc.
- Sends structured data to the backend for email generation.
-
Backend (FastAPI – Python)
- Handles API logic, context-aware email generation (LangGraph + OpenAI), and rate limiting.
- Caches responses and stores drafts for review.
-
Database (Supabase)
- Stores user data, email drafts, and application history.
- Uses SQL triggers to auto-update leaderboard points on each successful follow-up.
-
Electron App (Pet System)
- Gamifies progress: users adopt a digital pet that thrives only when they apply and follow up consistently.
- ✅ Real-time job detection & tracking
- 🧠 AI-generated personalized follow-ups
- 💬 Recruiter discovery via background service worker
- 🧩 Integrated leaderboard & gamified pet system
- 🗃️ Persistent data with Supabase + automatic point tracking
- 🧍 Fully functional local environment (frontend ↔ backend ↔ database)
- Implemented async message passing between background, content, and popup scripts via the Chrome runtime API.
- Solved delayed DOM rendering using a custom
waitFor()utility for Workday pages. - Integrated LangGraph to maintain contextual tone and company-specific personalization.
- Designed Supabase SQL triggers to automatically sync leaderboard points.
- Deploy backend via Render / Vercel and host data with Supabase.
- Add JWT-based user authentication.
- Implement caching + retry logic for API stability.
- Refine AI personalization logic for company-specific tone.
- Add tarot-reading side feature for user engagement (“fuzzies”).
- Publish Chrome extension for public testing.
Building QuickFollowup taught me how to:
- Structure cross-script communication across Chrome’s MV3 architecture.
- Manage async workflows between frontend, backend, and external APIs.
- Combine AI generation with real user workflows instead of toy demos.
- Ship MVPs fast and iterate calmly — even under complex, multi-system setups.
I’m Blair, a CS graduate passionate about building tools that blend AI + workflow automation to make people’s work easier and more fun.
Feel free to connect:
- 📧 Email: Jadepiper34@gmail.com