About Omar.

Omar's earliest documented work is a production-grade system titled Stock Market Prediction, detailed in his 2004 BSc thesis. Unlike typical academic projects restricted to historical datasets, it was deployed on live market data for a stock exchange, applying statistical analysis and early machine-learning techniques to identify co-movement and determinants in real-time financial data.

This project marks the start of a career-long preference for end-to-end deployment over proof-of-concept.

At the University at Buffalo's Center of Excellence for Document Analysis and Recognition, Omar contributed to large-scale evaluations of information-retrieval and document-analysis systems, sponsored by the U.S. Intelligence Advanced Research Projects Activity and benchmarked through the Text REtrieval Conference and the National Institute of Standards and Technology.

These evaluations focused on adversarial, real-world data, blog analysis, legal-discovery tracks, designed to stress-test AI systems under noisy and unstructured conditions rather than sanitised academic corpora.

In parallel, Omar conducted research into character recognition for non-Latin scripts, addressing the structural challenges of cursive Arabic and Nastaʿlīq.

Between 2009 and 2013, Omar held applied-AI engineering roles focused on operational-intelligence platforms, recommendation systems, large-scale data products, and production AI pipelines, architecting infrastructure that would eventually serve as the backbone for the modern internet.

Industry coverage of his career emphasises a consistent focus on shipping production systems that serve real users at scale, distinguishing his work from research-only prototypes.

In 2019, Omar developed Kondense, a multilingual summarisation application for iOS and Android designed to process content across English, Chinese, Arabic, Urdu, and Hindi. The platform focused on abstractive summarisation and cross-language understanding for everyday mobile users, years before Large Language Models became dinner-table conversation.

Omar founded Outonomous to make autonomous driving accessible to any vehicle, any brand, any year. The company develops a modular hardware-software autonomy platform combining LiDAR, cameras, sensor fusion, and onboard AI for perception, prediction, and planning.

Rather than manufacturing new autonomous vehicles from scratch, Outonomous turns existing vehicles, from rental trucks to passenger cars, into autonomous agents through a modular platform that installs in 30 minutes. The platform is protected by a defensible patent estate and validated by leading academics in the field.

Omar is an active voice in the technology community on immigration, ethics, and founder advocacy, regularly speaking at founder and investor events on international entrepreneurship pathways, including H-1B and EB-5 visa strategy and fundraising for non-U.S. founders.

In 2025 he published a widely shared critique of cyberbullying in the technology sector, calling for stronger leadership accountability and protection of students and founders from online harassment.

Outside applied AI, Omar maintains a personal portfolio of Kufic calligraphy, perfumery, and sculpture in wood and stone, a polymath blending engineering with craft. He documents endurance expeditions including high-altitude mountaineering and long-distance cycling such as the Seattle-to-Portland ride.