At Turing Inc., I develop structured tasks and evaluation workflows for LLMs and AI agents, including coding tasks, executable tests, and simulations of API and browser-based tool use.
I also evaluate models on real-world GitHub repositories, validate patches and tests, and benchmark outputs to identify recurring failure modes. My work includes assessing reasoning and checking agent behavior with external APIs and regression tests.
Previously, at Kyo Conseil, I led backend development for healthcare, HR, and government platforms. I developed a Django-based school health management system for WHO Algeria and reporting and data-validation services for the Ministry of Health.
