An autonomous full-stack monitoring system that tracks competitor web changes and generates strategic AI briefings using Gemini and FastAPI.
Abubakar Jibrin
I build AI-powered software, intelligent agents, RAG systems, automation tools, and modern full-stack applications.

Featured Projects
A responsive, client-side web application for real-time contact management and local browser persistence.
A Context-Augmented Generation (CAG) engine that leverages Gemini 2.5 Flash's long-context window for 100% recall Q&A and Pydantic schema extractions.
A full-stack, persistent financial logging application built with Python, Flask, and JSON storage engineered to process multi-byte local currency symbols across platforms.
A dual-tier number guessing application demonstrating cross-platform state management across Python terminal environments and modern browser interfaces.
A dual-interface security tool providing client-side high-entropy password generation across Python terminal and browser environments.
A production-grade developer tool that transforms rough human inputs into structured 5-component AI prompts using meta-prompting and side-by-side output evaluation.
A multi-tier calculator suite demonstrating clean separation of application logic across terminal, desktop GUI, and web-hosted interfaces.
A dual-interface assessment tool built with Python CLI and vanilla JS frontend featuring real-time evaluation and responsive card UI.
A responsive client-side task manager featuring custom CSS card layouts and persistent browser LocalStorage CRUD operations.
A grounded RAG system built with a custom NumPy cosine similarity vector store, gemini-embedding-2 vectors, and source-cited synthesis via Gemini 2.5 Flash.
About Me
I'm an independent software engineer focused on building practical AI-powered products. I work across AI agents, RAG systems, automation, APIs, and modern web applications, turning ideas into usable software.