Heidy Ibrahim

I’m an AI Language Evaluator and Indonesian & Javanese culture specialist, focused on delivering precise, guidelines-driven quality for language and AI datasets. I combine native cultural fluency with strong grammatical precision to ensure outputs are not only correct, but also locally appropriate.

In my current work, I evaluate Indonesian-language content for cultural appropriateness, grammar & search relevance, applying strict standards to maintain high-quality AI data labeling. I also support advertising standards through consistent evaluation practices that prioritize clarity, relevance, and factual integrity.

I’ve expanded into AI training and data preparation through preference ranking, prompt-response writing, and factuality evaluation. I deliver culturally relevant Indonesian language data to help improve conversational AI datasets, and I contribute hands-on quality checks that reflect RLHF-style evaluation and careful validation.

Alongside language evaluation, I annotate and label image datasets for AI model training, maintaining high inter-annotator agreement through rigorous quality standards and structured feedback. I’ve also built a multi-platform freelance track record and continue developing technical capability through Generative AI Developer training (including Python).