Data Quality · Evaluation · Reliable AI
I build reliable
language AI systems.
How I approach technical systems
Data before demos
Reliable systems start with understood data, explicit provenance, and documented quality constraints.
Evaluation before claims
A system is not useful simply because it produces an output. Its behavior must be measured, compared, and inspected.
Reliability by design
Testing, reproducibility, observability, and clear failure modes belong in the system from the beginning.
About
I am a Staff Data Engineer with a background in computational linguistics, applied NLP, language data, and technical leadership.
My academic work focused on linguistic data, annotation, modeling, and evaluation. In industry, I have worked across technical and linguistic initiatives and led multidisciplinary engineering teams.