The Operating Challenge
Searching for specialized roles across dozens of corporate career portals and job aggregators requires hours of manual browsing and filtering through irrelevant job posts.
Independent / Live System · Feb 2026 → Present01. Problem & Context 02. Architecture & Solution 03. Measurable Impact 04. Technology Stack
Automated Job Search & Match Engine
Self-improving custom tuned AI Agents monitoring 60+ platforms including job boards and enterprise ATS platforms, scoring role match via MPNet bi-encoders and Thompson-sampling feedback loops.
Technical Implementation
Built a 12-stage automated pipeline in Python 3.14 running 60+ custom tuned AI Agents. Features a dual-stage bi-encoder (all-mpnet-base-v2) and cross-encoder (ms-marco-MiniLM) scoring pipeline, coupled with a Thompson-sampling bandit learning engine that tunes filters based on user feedback.
Results & Verification
Surfaces top matched positions in a custom dashboard with 290+ pytest test cases verifying agents integrity.
Ask Bilal’s AI Clone