Top Machine Learning Jobs in Romania

Nebius seeks a Senior ML Engineer for AI R&D focusing on applied research in reinforcement learning and agent training, requiring deep expertise in machine learning and software engineering.

Design, build, and operate ML platform systems to train, evaluate, deploy, and serve generative models. Improve reliability, scalability, observability, and cost-efficiency of GPU and cloud workloads. Build internal tools and agent-friendly workflows, collaborate with researchers and engineers, and drive platform architecture and automation.

Design and build end-to-end ML systems for fraud detection: develop and deploy models, create production-ready features and real-time data pipelines, integrate with backend services, and ensure monitoring, security, and observability.

Design, build, and operate ML and data infrastructure on AWS. Migrate and optimize databases and ETL pipelines, build data lakes, process structured and unstructured data, develop and deploy ML and deep learning models (Jupyter), leverage AWS AI/ML services, and support customer-facing data solutions and migrations.

Lead a small senior data engineering team owning the end-to-end data platform: streaming, warehousing, modelling, pipelines, ML lift, and operational excellence. Drive delivery, coach engineers, review technical designs, enforce data contracts and quality, manage costs (FinOps), and partner with Product and leadership to set roadmaps and risk. Build and grow an emerging ML function.

Design, develop, and deploy ML models (classification, regression, NLP, CV, time-series). Prepare and feature-engineer data, build MLOps pipelines, containerize and serve models, monitor production performance, and collaborate with cross-functional teams to translate business problems into AI solutions while staying current on research and best practices.

Design, train, fine-tune, distill, and deploy LLM-based and agentic ML models. Build and operate agent workflows in Python, run evaluation and benchmarking, select models and cost strategies, and manage production monitoring and observability for model performance and reliability.

Own, design, build, and operate distributed, high-load backend services and data pipelines for ML infrastructure. Partner with ML and product teams to deliver self-service, reliable model serving, storage, and deployment components; ensure observability, scalability, and performance while driving technical decisions and continuous improvement.

Build and ship state-of-the-art speech systems end-to-end: model architecture, training (pre-train/fine-tune/post-train), evaluation, tooling, data curation, production inference, and safety/abuse mitigation for voice conversion and related tasks.

Build and operate production-grade model serving infrastructure, design deployment pipelines (blue/green, canary), implement autoscaling and multi-model serving, optimize GPU utilization and network throughput, set up observability and model registries, manage CI/CD for reproducible deployments, own full ML system lifecycle including on-call support and platform scalability.

Develop, deploy, and maintain machine learning models for operational defence and security systems. Work with data science teams to produce production-grade models using Python, SQL and ML frameworks, apply model deployment and MLOps practices, and integrate models into mission-critical systems.

Design, train, and improve computer-vision and generative ML models; handle end-to-end ML pipeline from data collection to deployment; prototype research ideas and convert successful prototypes into product-ready solutions; optimize models for efficient on-device inference and collaborate with product teams to integrate models.

The role involves developing AI solutions, managing the lifecycle of models, and collaborating with teams on computer vision projects. Candidates should have experience in AI and software engineering, with strong problem-solving skills.