Hi, I’m Gil!

I build AI systems that are meant to work in the real world.

I’m an early-career AI Software Engineer focused on applied AI systems such as ranking, personalization, evaluation workflows, and agentic behavior. My work spans system architecture, feature pipelines, and decision logic, with an emphasis on reliability, testability, and real-world performance.

I work on AI systems end to end, from architecture to evaluation.
I lead back-end development and system architecture for an AI-driven mobile app, designing data models, feed algorithms, preference systems, and analytics pipelines to support intelligent content ranking. I also work with Snorkel AI on LLM benchmarking, building evaluation tasks, reference implementations, and automated verification workflows in reproducible Dockerized environments.

Service and accountability shape how I approach engineering.
I’m a U.S. Armed Forces veteran and Chemical, Biological, Radiological & Nuclear (CBRN) Defense Specialist, trained in operational security, emergency response, and procedural compliance. These experiences reinforced the importance of reliability, attention to detail, and executing correctly when failure isn’t an option.

Resilience and perseverance have been core influences in my life.
My mother—an immigrant and single parent who lost her hearing in her 30s—built a successful career in software engineering without a college degree. From her, I learned adaptability, persistence, and the value of sustained effort.

I’m interested in AI as a system, not just a model.
I care about how AI behaves once it leaves the notebook—how it makes decisions, how it’s evaluated, and how it performs over time in real products. The projects in this portfolio reflect that philosophy.

I’m Gil Brandon Eskayo, and I’m excited to contribute my skills, perspective, and work ethic to teams building meaningful AI-driven systems.