JobSignal — Verify Before You Apply

Every day, job seekers waste hours applying to ghost jobs, fake listings, and positions that were filled weeks ago. JobSignal was built to fix that. You paste a job URL, a description, or a screenshot. The system then does the investigation: checking the company's official careers page, scanning LinkedIn, Indeed, Glassdoor, Reddit, and X for cross-platform signals, verifying the posting's age and freshness, detecting duplicate or recycled listings, and reviewing what current and former employees say about the company. The result is a clear, evidence-backed verdict: Apply, Verify, or Skip. Not a score you have to decode. Not a wall of technical data. A decision, with the reasoning behind it. JobSignal is honest about what it does not know. When evidence is thin, it says so instead of manufacturing confidence. The Verify verdict exists specifically for cases where the system found signals but not enough to commit either way. Key capabilities: - Single job verification via URL, pasted description, or screenshot upload - Batch mode for power users: paste up to 50 job links and get a ranked shortlist - Chrome extension that verifies listings inline while you browse, with no tab switching - Clipboard detection that auto-fills and triggers analysis in one click - Company reputation layer pulling from Glassdoor, Indeed, LinkedIn, Reddit, and X - A two-panel result view separating job verification signals from employer reputation - Step-based loading states that communicate what the system is checking in real time - Built with FastAPI, vanilla JS, Fireworks AI (Kimi K2) for language understanding, and Serper for live web intelligence The product philosophy has not changed from day one: accuracy first, then speed, then experience. But the past week proved those three things are not in conflict. Good UX is what makes accuracy useful.