AI Platform Engineer – OneAI

Role Overview

• Design, implement, and deploy advanced AI capabilities within the OneAI platform.
• Shape the end-user experience by designing intuitive workflows for model management, deployment configuration, and job operation.
• Streamline the model lifecycle by integrating public repositories (e.g., Hugging Face) for seamless discovery, import, versioning, and deployment.
• Bridge the gap between systems engineering and product design to ensure a seamless transition from backend infrastructure to user features.
• Integrate cutting-edge AI frameworks and engines, such as vLLM, NVIDIA Dynamo and Unsloth, into a secure and scalable environment.
• Leverage OpenNebula to orchestrate high-performance inference and training workloads across diverse cloud and edge environments.
• Develop and maintain reliable APIs for compute provisioning and workload scheduling.
• Implement GPU-aware operations to ensure optimal resource allocation and hardware utilization.
• Build comprehensive observability suites to monitor and track critical metrics, including latency, throughput, utilization, and failure rates.
• Establish and refine deployment and workflow strategies to ensure AI workloads remain efficient and stable at scale.
• Optimize system architecture to balance high performance with cost efficiency.
• Research and integrate emerging AI tools and engines to keep the OneAI platform at the forefront of the industry.
• Analyze performance bottlenecks to iterate on the efficiency of both training and inference processes.

Requirements

• Bachelor’s or Master’s degree in Computer Science, Information Technology, or Engineering.
• 3+ years of experience in applied AI, machine learning, or software engineering, with hands-on delivery of AI/ML solutions in production environments
• Demonstrated experience designing and deploying high-performance AI infrastructure, specifically focusing on the scalability and reliability of inference and training workloads.
• Proven track record of deploying Large Language Models (LLMs) at scale, with deep knowledge of serving engines (e.g., vLLM) and fine-tuning tools (e.g., Unsloth).
• Experience building AI-centric platforms or toolchains that manage the model lifecycle (versioning, deployment, and discovery).
• Experience with GPU orchestration and optimizing workloads for cloud, distributed or large-scale environments and collaborating with platform or infrastructure teams.
• Hands-on experience with high-throughput inference engines (e.g., vLLM) and fine-tuning tools (e.g., Unsloth)
• Proficiency in integrating with the Hugging Face ecosystem (Transformers, Hub, Datasets) for model and data management.
• Experience implementing monitoring tools to track system-level AI metrics such as token throughput, latency, GPU utilization, and failure rates.
• Experience designing and implementing scalable, reliable APIs for compute provisioning and workload scheduling.
• Experience working with cloud platforms and containerized environments (e.g., OpenNebula, Kubernetes)
• Advanced English level (B2 or higher) is required.

Tech Stack

• Cloud
• Kubernetes

Benefits

• Competitive compensation package and flexible remuneration: Meals, Transport, Nursery/Childcare
• Customized workstation (macOS, Windows, Linux)
• Private health insurance
• Paid time off: Holidays, Personal Time, Sick Time, Parental leave
• Afternoon-off working day every friday and during summer
• Remote company with bright HQ centrally located in Madrid; offices in Boston (USA), Brussels (Belgium) and Brno (Czech Republic); and access to office space near your location when needed.
• Healthy work-life balance: We encourage the right for Digital Disconnecting and promote harmony between employees personal and professional lives
• Flexible hiring options: Full Time/Part Time, Employee (Spain/USA) / Contractor (other locations)