I help teams build reliable data labeling systems, taxonomies, and data quality & evaluation workflows.
Working on something similar?
I work with startups and teams building AI products, especially those working with LLMs and human-labeled data.Ā Ā
Typically, I help with:
End-to-end data labeling/annotation workflows
Guidelines, QA, and evaluation frameworks
LLM + human hybrid processes
Build scalable classification systems
Normalize messy datasets
Improve search discoverability
Hands-on LLM-based data annotation workshops
Training teams on data quality & operations
Built and scaled data operations from thousands to hundreds of thousands of itemsĀ
Reduced manual labeling work by 60-80% using LLM workflows
Trusted to make high-stakes judgment calls in AI safety evaluations
Community lead at Data Operations IL (800+ members)
Regular speaker at industry conferences and meetups (you may have seen me at hayaData, Data TLV, PyData TLV, ××"××××, and more).Ā
