At DataAnnotation.tech, I evaluate and optimize Large Language Models for Hebrew accuracy, including syntax, grammar, idioms, and cultural context. I rate model responses for truthfulness, helpfulness, and safety using RLHF methodologies.
I write nuanced Hebrew prompts to test reasoning, translation, and multi-turn conversations, and document linguistic edge cases, semantic errors, and cultural biases in model outputs.
At Xiaomi | Israel, I provide technical troubleshooting for hardware, Android/MIUI software, and IoT devices. I also check device functionality before delivery and analyze customer feedback for recurring usability and software issues.
At IKEA | Israel, I conducted visual and functional QA/QC inspections on assembled products, documented defects, and made triage decisions based on quality benchmarks. I also tested products through hands-on assembly to check specifications, fit, and safety.
