Roles and moves.

Building AI systems at the core, in pursuit of sa/acc — where the work shapes what gets built nationally rather than at one company.

Built AI products and infrastructure at a Saudi AI startup, from the beginning.

ML systems for personal finance, data infrastructure for auditing, and backend features at a Saudi fintech.

Core member of the DARPA Machine Common Sense team, competing against Berkeley and MIT — first place in the majority of tracks. Designed a two-stage tracker combining appearance and motion cues, and helped build the large-scale internal evaluation system.

Machine Common Sense, on the Passive Violation of Expectation team under Dr. Alan Fern — detecting anomalies in object appearance and motion. Built the tracker, role assignment, and the reasoning agent that scored scene plausibility. Taught Python, databases, algorithms, networking, and graph theory.

Built and presented the Tech Data AI demo at IBM Think 2019, sponsored by Tech Data, IBM, NVIDIA and OpenPower. Deep learning projects on owl sound, plankton, and seed classification.

Founded Saudi Acceleration — building and shipping in public.

Non-profit backed by Misk Foundation & SDAIA — a data science & AI community.

Built Cursor Saudi — meetups and workshops across Riyadh and Khobar — and gave seven talks across the year.

Worked alongside the core maintainers on community-driven Arabic ML projects — Masader, the dataset catalogue, and Klaam for Arabic speech recognition and TTS — and mentored more than ten people through their first open-source contributions.

Ran it for six years — a developer community with a Discord server, open-source projects, and video content. It wound down the month sa/acc started.