LLM Research Engineer (NLP) - Remote Opportunity

About the Role

We are seeking an experienced LLM Research Engineer (NLP) remote to join our innovative team at Kyivstar.Tech. In this role, you will focus on large language models (LLMs) and cutting-edge AI research, contributing to the design and implementation of data preparation pipelines. Collaborating closely with data engineers, you will transform prototypes into scalable production pipelines and develop a state-of-the-art evaluation framework to measure and guide model quality.

What You'll Do

• Curate datasets for pre-training, supervised fine-tuning, and alignment of LLMs.
• Research and develop best practices and novel techniques in LLM training and evaluation pipelines.
• Collaborate with data engineers, annotators, linguists, and domain experts to scale data processes.
• Define evaluation tasks and collect high-quality feedback to improve model performance.
• Engage in end-to-end LLM training, ensuring models are aligned with user needs and ethical standards.

Requirements

• 3+ years of experience in Data Science or Machine Learning, with a focus on NLP.
• An advanced degree (Master’s or PhD) in Computer Science, Computational Linguistics, or a related field is highly preferred.
• Practical experience with fine-tuning LLMs and modern NLP approaches.
• Strong experience with deep learning frameworks such as PyTorch or JAX.
• Hands-on experience with containerization (Docker) and orchestration (Kubernetes) for ML.

Nice to Have

• Experience with Reinforcement Learning in NLP settings.
• Prior work on LLM safety, fairness, and bias mitigation.
• Publications in NLP/ML conferences or contributions to open-source NLP projects.

What We Offer

• Flexible work environment: office or remote — it’s up to you.
• Remote onboarding and training opportunities through internal resources.
• Health and life insurance, along with a wellbeing program.
• Reimbursement of expenses for Kyivstar mobile communication.
• Supportive work culture that values innovation and continuous learning.