Yasin Yaqoobi
Afghan-American engineer and applied-AI lead, Northern Virginia
He leads an applied-AI lab at a Fortune 500 company, and started Partaw to bring that work to Afghan healthcare.
- Role
- Founder; builds and operates the platform
- Field
- Applied AI: search, model training, inference, agents
- Languages
- Dari, Pashto, English, Urdu
- Based
- Northern Virginia, United States
- Education
- University of Arizona, Eller College of Management — management information systems
A film nobody could read
Partaw began with a plain observation: Afghan clinics could take X-rays that nobody was in a position to read. The machinery to fix that already existed in everyone’s hands — a phone, a messaging app, and a distant radiologist willing to look.
What was missing was the part in between: something that takes a photographed film from a clinic’s phone, puts it in front of a licensed radiologist in a form they can read, and carries a signed report back in the clinic’s own language. That part is software, and it is what he builds.
Applied AI, in production
Yasin leads the Applied AI Lab at a Fortune 500 company, where he is the person teams across the business turn to on AI. The lab’s work has reshaped the company’s consumer-facing product, most visibly a natural-language search that lets people describe what they want in their own words and get it.
The work runs the full depth of the field. He has designed and trained neural network models himself, and wrote the standards the organization follows for model inference and fine-tuning. His team builds AI systems that span departments: enterprise platforms, autonomous agents, and deep-research agents that read, reason, and report across the company’s own knowledge. More than once the lab has had an approach running in production before it appeared in published research or in commercial tooling.
By Afghans, for Afghans
A native speaker of Dari and Pashto, he studied management information systems at the University of Arizona’s Eller College of Management on international scholarships from the university and the Aga Khan Foundation, and served as president of the university’s Afghan Student Association.
His first project for Afghans came long before Partaw: daridictionary.org, the first comprehensive English-to-Dari dictionary. It set the pattern he has kept since — durable tools built by Afghans, for Afghans, that keep working without outside help.
What he wants AI to do for Afghan healthcare
Afghanistan has too few specialists and too little money for care. AI can stretch both: lower the cost of a diagnosis, and let one trained reader serve many more patients. Partaw is where he puts that to work.
- Outbreak detection
- Turning everyday reads into an earlydisease signal that national surveillance does not have today.
- Standard processes, standard data
- Giving clinics a common way to request, record, and report a study, so medical data can be compared from one province to the next.
- Protection against counterfeits
- Mohr-e-Aseel (مهر اصیل), the seal that lets anyone check a medical document is genuine.