Firki
Agentic AI that sees and extracts technical signals from JD
2 followers
Agentic AI that sees and extracts technical signals from JD
2 followers
Firki is the first Agentic Vision Model (AVM) built to solve the 73% "Keyword Gap" in talent discovery. While traditional ATS systems rely on rigid text matching, Firki’s AI "sees" unstructured JDs in Gmail/Outlook to infer true technical intent. Key Impacts: 97.5% Efficiency: Reduces Boolean building from 20 mins to 30 secs. Ethical AI: Focuses on skill signals over pedigree to reduce bias. ROI: Saves recruiters $3k+/mo in manual labor.
Hi Product Hunt! 👋
I’m Pankaj, and I’m incredibly excited to share Firki with you all today.
After years of leading global engineering teams as VP of Technology at Rocket and consulting for NVIDIA, I saw a persistent, trillion-dollar problem: the 'Keyword Gap.' Traditional recruitment tools are rigid—if a candidate doesn’t use the exact words a recruiter is looking for, they are effectively invisible. This failure results in a 30%+ loss of qualified talent and perpetuates systemic hiring bias.
I built Firki independently to fix this using Agentic Vision Models (AVMs).
Why Firki is a "Game-Changer":
Beyond Keywords: Our AVM architecture doesn't just parse text; it 'sees' and reasons through job descriptions to infer technical intent.
Massive Efficiency: We’ve seen recruiters reduce their search-prep time from 20 minutes to just 30 seconds (a 97.5% gain).
Ethical by Design: By focusing purely on technical signals and removing 'pedigree' noise, Firki promotes a truly skill-based, unbiased hiring process.
This isn't just about speed; it’s about making the global talent market more equitable. Firki is my contribution to the next era of Agentic AI.
I’d love to hear your thoughts! What are the biggest 'keyword' headaches you face in your workflow?
I’ll be here all day to answer your questions. Let’s build a smarter, fairer future for hiring together! 🚀"