ML/NLP Engineer (AI)

TL;DR

ML/NLP Engineer (AI): Developing and optimizing advanced machine learning and natural language processing solutions with an accent on end-to-end ML pipelines and production readiness. Focus on implementing LLM and RAG pipelines, fine-tuning models for business impact, and adopting MLOps best practices.

Location: Onsite in Novi Beograd, Serbia

Company

A fully Open Access publisher with a portfolio of more than 500 journals across all scientific disciplines, developing most of its software in-house.

What you will do

• Develop, optimize, and evaluate ML and statistical models tailored to specific business needs.
• Design and deploy scalable NLP models, ensuring seamless integration into existing production systems.
• Oversee the entire ML pipeline from data collection and preprocessing to training and deployment.
• Implement monitoring and maintenance strategies to ensure consistent model performance over time.
• Build data pipelines and interactive prototypes using Streamlit to gain insights from complex datasets.
• Collaborate with data scientists, software engineers, and product managers to align AI initiatives with business goals.

Requirements

• Master’s degree in Data Science, ML, or a related field.
• Over 3 years of experience in ML/NLP Engineering.
• Advanced proficiency in Python for complex applications and ML models.
• Hands-on experience with PyTorch, TensorFlow, Scikit-learn, and the Hugging Face ecosystem.
• Strong expertise in NLP techniques, including Tokenization and Named Entity Recognition (NER).
• Solid experience with FastAPI, Celery, Large Language Models (LLMs), and RAG pipelines.

Nice to have

• Additional MLOps and AI/Data Science experience.
• Familiarity with MLflow and Kubeflow.
• Proficiency using Pandas and Polars.

Culture & Benefits

• Opportunity to work at the forefront of digital transformation in scientific publishing.
• Technology-driven environment using modern open-source stacks.
• In-house software development approach promoting creativity and technical ownership.
• Collaborative culture focused on knowledge sharing and contributing to the scientific community.