sam obara

I’m a Data Analyst and AI/ML Specialist with 5+ years of hands-on experience building, evaluating, and improving machine learning systems in fully remote environments. I focus on structured data analysis, ML model evaluation, and prompt engineering to measurably improve model accuracy and data quality at scale.

At DataFlow Solutions, I designed and executed structured data annotation pipelines for supervised ML models—labeling 10,000+ data points monthly with high precision across text, behavioral, and categorical datasets. I led QA testing on production AI outputs, identifying systematic failure modes and edge cases to drive a 25% improvement in classification accuracy, while building evaluation rubrics and scoring frameworks to benchmark performance across multiple task types. I also documented training procedures and annotation/evaluation protocols to reduce onboarding time for new analysts, and I collaborated with engineering teams in Agile sprints to integrate feedback into retraining cycles.

Previously, at TechVerify Global, I delivered human-in-the-loop feedback to improve precision and recall across content categories, processing 500+ data cases daily while sustaining a 98% quality and accuracy score. I analyzed misprediction patterns and produced structured insight reports for model retraining, supported weekly global calibration for consistency, and worked with AI-assisted moderation tooling to flag systematic biases. I began as a Virtual Research Analyst, gathering and cleaning data from research databases and APIs and translating findings into clear, data-driven recommendations—skills that still guide how I communicate results and drive continuous improvement.