Full-stack developer
I build clean web apps, mobile experiences, and AI-powered tools.
Focused on React, React Native, TypeScript, APIs, and practical AI systems.
Selected work
HYSA and Souji are the center of the portfolio: AI tooling, mobile education, APIs, and clean product UX.
Core product / AI coding assistant
An AI coding assistant with local project context, provider fallback, attachments, and reviewable edits.
Core product / educational app
A mobile-first study app for Algerian students with grade tools, saved resources, and AI help.
Skill map
The emphasis is not novelty for its own sake. It is the combination of product thinking, typed implementation, AI capability, and interface craft.
Responsive interfaces with disciplined component boundaries, readable state, and motion that supports the workflow.
Typed API layers, database modeling, auth middleware, upload services, and response contracts built for production paths.
Cross-platform app flows with reusable primitives, fallback states, theme context, and RTL-aware layouts.
Multi-provider model systems with task routing, diagnostics, streaming, vision handling, and safe user control.
Clear contracts for product features that need data, files, identity, rate limits, and reliable fallback behavior.
Product surfaces that stay quiet, sharp, readable, and believable under repeated use.
About
Yahia's work is shaped by practical constraints: local developer workflows, student education needs, provider instability, RTL mobile usability, and codebases that need to stay understandable.
I build products from the workflow outward. HYSA Code started as a local AI coding assistant, then grew into a terminal and browser workflow with project context, file tools, provider routing, attachments, and safe edit controls.
Souji AI applies the same product mindset to education: a mobile-first study platform for Algerian students with grade calculation, souji library flows, saved resources, admin publishing, and AI-assisted explanations.
My engineering style favors typed boundaries, reusable interface primitives, fallback-aware AI, RTL-ready mobile UX, and code paths that stay understandable after the first demo.
01
Product clarity before feature volume
02
Typed architecture that can survive growth
03
Interfaces that remove hesitation and visual noise
04
AI integrations with fallback paths and user control
Contact
The fastest fit is a project that needs thoughtful frontend execution, a reliable TypeScript backend, or an AI workflow that has to work outside a demo.
For collaboration, freelance work, or product discussions, reach out through any of the channels below.
