Инженер инфраструктуры машинного обучения (Middle MLOps Engineer)

Location

Madrid, Spain, or remote across Europe and the United Kingdom; applicants must be authorized to work in the country where they apply.

About the company

A global AI cloud provider building a full-stack platform that takes developers and enterprises from data and model training to production deployment. Its 1,500+ person team, including hundreds of engineers, works across compute, storage, networking and applied AI, with R&D hubs in Europe, the UK, North America and Israel.

Responsibilities

• Tune GPU clusters and InfiniBand networks to keep HPC and GPU workloads running at peak performance.
• Diagnose the root causes of GPU and InfiniBand issues and drive corrective fixes.
• Bring new hardware, including new GPUs, into the existing stack via Kubernetes, QEMU and KVM.
• Build out automation for proactive monitoring, fault detection and resolution, and manage GPU devices and InfiniBand fabrics.

Requirements

• 5+ years of system-level software development focused on performance optimization and low-level programming.
• 3+ years of hands-on Linux administration, troubleshooting and performance tuning.
• Solid grasp of server architecture: PCIe devices, NICs, the Linux kernel and HPC systems.
• Strong skills in at least one performance-oriented language such as C/C++, Go or Python.

Location

The role is based in Lithuania, with compensation for other locations discussed during the interview process; no specific remote or on-site format is stated.

About the company

A global financial technology company building a broad consumer finance super app, serving tens of millions of customers with products spanning spending, saving, investing, exchanging, and travel.

The technology organization develops the infrastructure and features behind the app and is expanding its internal AI agent capabilities within a heavily regulated financial environment.

Responsibilities

• Design, build, and maintain scalable Python services, APIs, and data pipelines for an internal agent platform
• Develop and deploy LLM agents, covering tool integrations, orchestration, and state management
• Create reusable platform capabilities so teams can develop, test, and run AI agents in production
• Improve agent reliability, observability, performance, and cost efficiency through evaluation, monitoring, and robust error handling
• Break down complex AI and backend problems into technical designs, reviewable components, and delivery plans, and review designs and code for security and maintainability

Requirements

• 5+ years as a software engineer focused on backend systems, including 3+ years using Python as the primary language
• Hands-on experience building LLM agents, with production experience in AI agent systems or platforms, plus a STEM academic background
• Fluency in SQL and proficiency in another object-oriented language, with experience in API development, integration, and distributed systems
• Ability to write concurrent code for I/O-bound and CPU-bound workloads and familiarity with Docker, Kubernetes, CI/CD, monitoring, and alerting
• Understanding of production agent challenges such as tool execution, context management, evaluation, and failure handling, and the ability to turn ambiguous requirements into clear designs and trade-offs