JustHireMe  by vasu-devs

Local-first AI job intelligence workbench

Created 2 months ago
1,895 stars

Top 22.5% on SourcePulse

GitHubView on GitHub
Project Summary

This open-source desktop workbench addresses the noise and opacity of traditional job boards and AI application tools. It empowers job seekers, particularly those in technical AI roles, by providing a local-first system for discovering relevant opportunities, transparently evaluating fit, and generating customized application materials. The project offers greater control and signal compared to cloud-first, black-box solutions.

How It Works

JustHireMe operates as a local-first AI job intelligence workbench. It scrapes leads from diverse sources, applies a deterministic quality gate to filter out low-value postings, and ranks remaining opportunities based on explainable rules, feedback learning, and optional LLM reasoning. Candidate fit is assessed using Kuzu graph data and LanceDB vectors for semantic matching against a user's profile. Finally, it customizes application materials like resumes, cover letters, and outreach drafts. This approach prioritizes transparency, local data control, and extensibility for contributors.

Quick Start & Requirements

  • Primary install/run:
    • Frontend Preview: git clone the repository, cd JustHireMe, npm install, then npm run dev.
    • Full Desktop Setup: git clone the repository, cd JustHireMe, npm install, cd backend, uv sync --dev, cd .., then npm run tauri dev.
  • Prerequisites: Node.js (20+), Python (3.13+), Rust (stable), uv (latest stable), Git. Optional: Ollama for local models, Playwright for experimental automation.
  • Documentation: Architecture (docs/ARCHITECTURE.md), Source Adapters (docs/source-adapters.md), Contribution Guide (CONTRIBUTING.md), Roadmap (ROADMAP.md).

Highlighted Details

  • Scrapes job leads from a wide array of sources including ATS, feeds, communities, and APIs, preventing vendor lock-in.
  • Features a deterministic quality gate to filter out stale, spammy, or low-context job postings before they pollute the pipeline.
  • Provides explainable ranking that details why a job is considered a good fit, incorporating profile context via Kuzu and LanceDB.
  • Generates tailored application packages: resume PDFs, cover letter PDFs, founder messages, LinkedIn notes, and cold emails.
  • Employs a local-first architecture, storing profile data, lead history, and generated documents on the user's machine by default.

Maintenance & Community

The project is maintained by Vasu-devs and is currently in alpha. Contributions are welcomed for source adapters, ranking improvements, and packaging. The maintainer is open to sponsorship to support ongoing development and is reachable via email or X/Twitter.

Licensing & Compatibility

JustHireMe is released under the "JustHireMe Personal Open Source License." This license permits personal, educational, research, and evaluation use. However, all forms of commercial use, including business operations, paid services, lead generation, or any direct/indirect financial benefit, are strictly prohibited without express prior written permission and a separate royalty agreement.

Limitations & Caveats

The project is in alpha and not yet a polished consumer product. Browser automation and auto-apply features are experimental, unsupported, and disabled by default. The restrictive "Personal Open Source License" significantly limits commercial adoption and integration into business workflows. Windows desktop packaging is the primary release target.

Health Check
Last Commit

2 days ago

Responsiveness

Inactive

Pull Requests (30d)
65
Issues (30d)
24
Star History
1,908 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.