Most companies already use AI. It just lives in a dozen private accounts. I turn that fragmented, individual use into one governed, company-wide AI platform: AI agents now, and private local AI on the company's own hardware next.
When I joined HASIB in October 2026 to lead its AI transformation, AI was not new there. People were using it every day.
But each person used it individually, in their own account, with their own tools and habits.
Governance, knowledge and cost all sat with individuals. A shared rule for what data goes where, a place to share what works, and one view of cost were the natural next step.
So I proposed one company-wide AI agent platform. Each department gets agents shaped to its own work, instead of each person working out their own setup.
Access rules decide who can use which agent. Data rules decide what information each agent may touch, and what never leaves its department.
Every agent reports into one central place, so the company can see how AI is used and what it costs. That's the platform I'm rolling out now.
The next step is private, local AI, running on the company's own machines. The same agents and the same rules, with company data kept on company machines.
I've run my own product this way for years. أسئلة ثقافية is built by a team of AI agents, and a human still signs off every release.
Cultural Questions is an Arabic trivia game I founded in 2011. It started on React Native, PHP and MySQL, moved to Flutter and Firebase, and is now native Android with a cloud backend, crash monitoring and analytics.
Most of it plays offline. Online, there are duels, weekly leagues, a Countries Cup, friends and a daily challenge. Version 4.5, in progress, adds a One Thousand and One Nights story game and a city builder.
Private local AI means the models run on hardware the company owns, under the same access and data rules as the agents. I'm planning it end to end.
Leads the company's AI transformation. Proposed and is rolling out a company-wide AI agent platform: agents per department, access and data rules, a central view of usage and cost. Planning private local AI: hardware sizing, model choice, deployment and management.
Arabic trivia game with 2M+ users and a 4.5★ rating from 9,400+ reviews. 77,000+ questions with explanations across about 200 topics and 46 games. Built with coding, writing, fact-checking and art agents, with human review before release.
On a platform used by 50+ travel companies: a monthly IP-range data pipeline to detect and block fraud (.NET MVC, C#, SQL), a product customisation system (.NET MVC, C#, LINQ) and custom SSRS reports for large clients. Fixed 100+ complex bugs; top of the weekly bug leaderboard.
Data warehouse on Microsoft BI, SSRS reports and dashboards, and an SSAS tabular finance datamart.
HR IT on SAP at Saudi Aramco (2011); a finance dashboard on SAP BusinessObjects at Savola Group (2012). Appreciation certificates from both.
AI twin, can make mistakes. For anything important, email the real Taha.