open-job-hunter discovers job offers from multiple boards and scores each one against your configurable profile using a local LLM. No cloud. No data leaving your computer.
One profile.json drives everything. Describe your stack, seniority and targets — the tool works for anyone.
Every offer gets a 0–100 match score, apply decision, risks, strengths and a tailored pitch angle — in Spanish or English, via your own Ollama model.
Pulls openings from Remotive, Greenhouse and Lever boards out of the box, plus generic URL ingestion.
Search, sort and page through large local histories, with separate views for active applications, received offers, rejections and discarded roles.
Optionally run discovery on a schedule and get the top matches emailed to you.
Evaluation runs on a local model. Your profile and pipeline never leave your machine.
Two interfaces — a standard MCP server for Claude and any MCP client, and an HTTP bridge for dashboards.
MCP client (Claude, etc.) + HTTP dashboard
↓
open-job-hunter — discovery · evaluation · pipeline
↓ prompt built from your profile.json
Ollama (local LLM) — 🔒 your data stays here
Requires Node ≥ 20 and a running Ollama instance.
# 1. Clone & install git clone https://github.com/andminin-engineering/open-job-hunter.git cd open-job-hunter && npm install # 2. Pull a local model ollama pull qwen2.5:7b # 3. Make the profile yours cp config/profile.example.json config/profile.json # ...edit config/profile.json with your background # 4. Build & run npm start
This project started as a personal job-search tool and was generalized so anyone can use it. Contributions that make it useful to more people are especially welcome.