About
Senior ML Engineer and hands-on Technical Lead working on forecasting, retrieval and applied AI.
I’m Hasan Hüseyin Yurdagül, a Senior ML Engineer and hands-on Technical Lead working on forecasting, retrieval and applied AI. I design and ship machine learning systems where wrong predictions have real costs — model development, evaluation, and MLOps, from pilot to production.
I’m currently leading the AI R&D roadmap at Universal Software, mentoring engineers and working on forecasting and support intelligence. Alongside that role, I consult for Moka on fintech models and their serving pipelines.
Consulting
My work starts with the decision a model should support, then a data audit: what the target means, what information is available at prediction time, and whether the data supports the promise. From there I compare models, communicate the results, and package the useful parts as reports, apps, or APIs.
That has meant connecting call-volume forecasts to Erlang staffing calculations at Ikas, building search and delivery analytics at Inveon, and implementing model-specific feature contracts for batch inference at Moka. Mentoring is part of the handoff too: two Albert Solino engineers built their own Flask version of MLStudio, and three Gtech data scientists built a desktop one with my guidance. The work page has the roles, engagement dates, and tech stacks; Moka is my only active consultancy.
Background
- BSc in Computer Engineering (English-medium), Çukurova University, Sep 2016–Aug 2021.
- MSc in Computer Engineering, Çukurova University, Aug 2023–Present · thesis in progress.
- Thesis: AI-Driven Object Size Prediction in Visual Data Without Depth Information and Reference Points. Fishometry estimates fish length from a single uncalibrated photo, comparing geometric, relative-depth, and learned image features.
- Applied work across forecasting, computer vision, recommenders, and NLP/LLM systems, including serving and model handoff.
- I write up what I learn in articles.
Selected Publications
- Fault Prediction in Energy Systems: Telemetric Data and Machine Learning Approaches — IEEE ISAS, 2025; first author.
- Development of Machine Learning-based Revenue Forecasting Models for Electronic Payment and Money Institutions — IEEE ISMSIT, 2025; co-author.
- Forecasting Call Center Arrivals Using Machine Learning — Osmaniye Korkut Ata University Journal, 2021; co-author.
Contact
The fastest way to reach me is email. Tell me about the decision you want a model to support.