All jobs
Data Scientist
Kaizo
Amsterdam · HybridEntry level (0–2 years)Full-timePosted 14 Jul 2026Visa sponsorship
About the job
What you'll do
Own the evaluation and quality loop of the AutoQA product by translating customer rubrics into testable LLM instructions. Design and execute experiments to improve accuracy and build internal LLM-as-a-judge pipelines for repeatable assessment.
What they're looking for
Requires 0-2 years of experience with a strong understanding of LLMs, applied statistics, and Python. Candidates should possess an analytical mindset and the ability to communicate findings to both technical teams and customers.
Full description