ML Engineer (NLP / LLM)

About the Position

ML Engineer (NLP / LLM)

We are looking for an ML Engineer with solid hands-on experience in Natural Language Processing, Large Language Models, and fine-tuning models on client-specific data.

This role is ideal for someone who enjoys solving real business challenges with modern AI tools and can turn raw client data into production-ready intelligent solutions.

You will join a project focused on building custom AI features powered by LLMs, including data preparation, model adaptation, pipeline development, and deployment support.

Responsibilities

• Fine-tune and adapt LLMs such as GPT, LLaMA, and similar models for domain-specific use cases
• Prepare, clean, normalize, and structure raw text data provided by clients
• Build scalable NLP pipelines for training, inference, and integration into real products
• Work with tools and frameworks such as PyTorch, TensorFlow, Transformers, Hugging Face, and LangChain
• Select and integrate appropriate tools for embeddings, vector search, model serving, and inference optimization
• Collaborate with developers, product teams, and stakeholders to deliver AI-driven features aligned with business needs

Requirements

• Practical experience with LLM fine-tuning and prompt engineering
• Strong knowledge of PyTorch and/or TensorFlow
• Experience with LangChain and NLP pipeline composition
• Good understanding of text preprocessing, normalization, and dataset preparation
• Experience deploying ML/NLP models in production environments
• Knowledge of vector databases, embeddings, retrieval pipelines, and inference optimization
• Ability to work independently and propose implementation approaches based on project goals

Cooperation Format

• Project-based engagement
• Opportunity to work on cutting-edge AI solutions for real client cases
• Close collaboration with an experienced delivery team
• Flexible setup depending on project scope and availability

Apply Now

Interested candidates are encouraged to apply and join our innovative team.