London, UK

Amr Hasanin

I build AI systems and apps for real clients, and I study Applied AI at Queen Mary University of London.

Email me System AI

About

I am a second-year BSc Applied Artificial Intelligence student at Queen Mary University of London, in the programme's first ever cohort.

I also run System AI, a small studio that builds software for businesses in the UK, Australia, Saudi Arabia and the US. I write Python, TypeScript and Kotlin. I put AI agents into production with approval gates and checks on every answer.

Selected work

Appointed Duty

Android and iOS prayer alarm app

An alarm that rings before each congregational prayer and asks for a short mission to switch it off. Flutter with a native Kotlin alarm service on Android. More than 600 automated tests, and a GitHub Actions workflow that runs them and builds a signed release on every push. Live with real testers on Google Play.

Flutter, Kotlin, GitHub Actions

The Srama Group

WhatsApp to CRM automation

An AI intake pipeline for a 12-person Australian real estate agency. It transcribes voice notes, works out what each message needs and routes it into the CRM. Admin work fell by 97%, about 84 staff-hours a week. The agency won Queensland Agency of the Year.

n8n, Zoho CRM, speech to text

Kestrl

Stock screener for a UK fintech

A Sharia-compliant stock screener, built from concept to production in 7 days and live on the Kestrl app.

Web app, financial data

tus-kahoot

Quiz app

A Kahoot-style quiz app built for a medical student preparing for the Turkish licensing exam. Strict TypeScript front and back end, a Supabase database, and PDF export of question sets.

Next.js, TypeScript, Supabase

AI-OS

My own operating system

A Python system I use every day: a Telegram bridge, scheduled background agents that draft, file and remind, Supabase storage, and offline test suites with several hundred checks. Live since July 2026.

Python, agents, Supabase

Data projects

University work

A weather-aware regression that forecasts UK flight delays from CAA and METAR data (R² 0.782, 82% monthly accuracy on unseen data). A composite index of graduate outcomes for 153 UK universities, with salary predictions 93% accurate.

Python, scikit-learn, pandas

Experience

Education

Skills

Languages
Python, TypeScript, JavaScript, Kotlin, Dart, SQL
Build and ship
Flutter, React and Next.js, Postgres and Supabase, REST APIs and webhooks, Docker, Git, GitHub Actions, Google Cloud
AI systems
LLM APIs, agents and tool calling, evaluation of model answers, local open models with Ollama, retrieval-augmented generation, LangChain, n8n
Spoken
English (fluent), Arabic (native)

Contact

The best way to reach me is email.