AI Researcher (Multilingual Data)

TL;DR

AI Researcher (NLP/ML): Building and scaling next-generation language models across diverse languages and domains with an accent on data sourcing, curation, and training strategies for multilingual and low-resource languages. Focus on publishing high-quality research at top venues and translating these insights into production-ready systems.

Location: Remote (World)

Company

hirify.global is a fast-moving startup building frontier language models with a focus on high-quality research and real-world impact.

What you will do

• Design and execute research on multilingual datasets, including collection, filtering, deduplication, and quality measurement.
• Develop augmentation, sampling, and curriculum design strategies for low-resource and long-tail languages.
• Improve cross-lingual transfer, alignment, and robustness in large language models.
• Build and maintain evaluation benchmarks for multilingual performance.
• Collaborate with engineers on model architecture decisions and training pipelines.
• Publish research at top venues such as ACL, EMNLP, NeurIPS, ICML, and ICLR.

Requirements

• Strong background in NLP/ML research with a focus on multilingual or cross-lingual modeling.
• Proven publication record at respected conferences or journals (e.g., NeurIPS, ICML, ICLR).
• Experience working with large-scale text datasets across multiple languages.
• Deep understanding of tokenization, vocabulary design, and data quality metrics.
• Proficiency in Python with modern ML frameworks such as PyTorch or JAX.
• Ability to operate independently and ship results at a startup pace.

Nice to have

• Experience with non-Latin scripts or low-resource languages.
• Contributions to open-source NLP or data tooling.
• Direct experience in training or evaluating large language models.
• Familiarity with multilingual benchmarks like XTREME, FLORES, or TyDi QA.

Culture & Benefits

• Direct ownership over research direction and product impact.
• Culture that equally values academic publishing and production deployment.
• Access to large-scale datasets and modern infrastructure for fast iteration.
• Competitive compensation and early-stage equity.