Inference Platform Engineer (LLM & Kubernetes)

N-iX is a global software development service company that helps businesses across the globe create next-generation software products. Founded in 2002, we unite 2,400+ tech-savvy professionals across 40+ countries, working on impactful projects for industry leaders and Fortune 500 companies. Our expertise spans cloud, data, AI/ML, embedded software, IoT, and more, driving digital transformation across finance, manufacturing, telecom, healthcare, and other industries. Join N-iX and become part of a team where your ideas make a real impact.

We are looking for an Inference Platform Engineer (LLM & Kubernetes) to join our team.

Our client is a leading European AI company developing large language models and generative platforms for enterprise and government clients.
Their products combine high-performance technologies, transparency, accessibility, and data security, fully aligned with European regulatory and ethical standards.

As an Inference Platform Engineer (LLM & Kubernetes), you will take ownership of inference API integration, operations, and platform reliability across production AI systems.
This role is designed to be covered by 1–2 FTE split across several senior specialists, ensuring continuity of inference services and full coverage during planned and unplanned absences as we take over end-to-end LLM inference responsibility.

Responsibilities:

• Take ownership of inference API integration, orchestration, and long-term platform reliability


• Lead operations for LLM inference services as they transition under internal ownership


• Ensure inference API availability, latency, and performance in production environments


• Design and maintain multi-turn conversation handling, chat templates, and prompt orchestration


• Proactively monitor, troubleshoot, and resolve inference platform issues, logs, and errors


• Manage Kubernetes deployments, Helm charts, and ArgoCD workflows for inference services


• Ensure platform security, CVE monitoring, and compliance with internal and regulatory standards


• Collaborate closely with backend, platform, and infrastructure teams


• Maintain clear operational documentation to support shared ownership across multiple FTEs

Requirements:

• 5+ years of Python programming experience


• Strong Kubernetes (k8s) experience, including deployment, scaling, and monitoring


• Experience handling large-scale logs, monitoring, and observability in production


• Basic knowledge of LLM fundamentals and the surrounding industry  (e.g., what type of models exist, how does an LLM generate output)


• Experience from the user side developing against an Inference API (e.g., OpenAI, Anthropic, OpenRouter etc.) and understanding  of their structure (experience with providing or deploying a similar API yourself a strong plus)


• Ability to independently own and operate inference services in a shared-responsibility model (1–2 FTE split across multiple specialists)


• Strong communication skills and experience working with cross-functional engineering teams


• Solid Linux fundamentals

Nice to have:

• Hands-on experience with Helm charts, ArgoCD, and CI/CD for AI services


• Interest in partly working with Rust


• Senior-level experience with production LLM inference or AI platform operations


• Experience building or operating multi-turn conversational AI systems


• Familiarity with real-time API orchestration or streaming inference workloads


• Background in MLOps, AI platform engineering, or SRE


• Experience with cloud-based inference deployments and scaling


• Knowledge of security, CVE scanning, and operational best practices

Technology Stack:

• Inference: OpenAI, Anthropic, or other LLM inference APIs


• Focus Areas: API integration, multi-turn conversation orchestration, tool calling, platform reliability


• Infrastructure: Kubernetes, Helm, ArgoCD, cloud or hybrid environments


• Monitoring: Logs, metrics, observability tools for inference systems


• Workflow: Git, CI/CD pipelines, documentation, operational runbooks, incident handling


• Standards: Reliability, latency, performance, security, maintainability