Senior AI Platform Engineer — Scalable GPU-Driven Agents

Singularity IO GmbH; Zürich; Hybrid; CHF 150,000 - 230,000; Full time

Job summary

Singularity IO GmbH in Zürich seeks a Senior AI Platform Engineer to own the sovereign AI platform infrastructure, designing and optimizing GPU-accelerated runtime, ensuring reliability, compliance (EU AI Act, GDPR), and scalable performance for clients and internal digital workforce.

You will build Kubernetes-based pipelines, optimize Ollama inference and vector store (Qdrant), manage cost-efficient GPU scheduling, and collaborate with engineers to enable smooth transitions across workflows.

Qualifications

• 5+ years of experience in AI/ML platform engineering or MLOps.
• Strong expertise with Kubernetes, GPU acceleration, and high-performance inference.
• Hands-on experience with Ollama, vector databases (Qdrant), and containerized LLM serving.
• Proficiency in Python, Terraform, and infrastructure-as-code practices.
• Deep understanding of cloud cost optimization and performance tuning at scale.
• Experience in regulated environments (EU AI Act / GDPR) is a strong advantage.

Responsibilities

• Design and maintain the core sovereign platform running on Exoscale GPU clusters
• Optimize Ollama inference, model serving, and vector store performance (Qdrant)
• Build robust Kubernetes-based deployment pipelines and observability systems
• Implement cost-efficient GPU resource scheduling and auto-scaling strategies
• Ensure full EU AI Act and GDPR compliance across the entire platform layer
• Collaborate with Agentic AI Engineers to enable seamless transition from Dify low-code to LangGraph production workflows

Skills

Kubernetes

GPU acceleration

High-performance inference

Ollama

Qdrant

LLM serving

Python

Terraform

IaC

Cost optimization

EU AI Act knowledge

Tools

Ollama

Qdrant

NVIDIA GPU Operator

LangGraph

Dify

n8n

Job description

Singularity IO GmbH in Zürich seeks a Senior AI Platform Engineer to own the sovereign AI platform infrastructure, designing and optimizing GPU-accelerated runtime, ensuring reliability, compliance (EU AI Act, GDPR), and scalable performance for clients and internal digital workforce.

You will build Kubernetes-based pipelines, optimize Ollama inference and vector store (Qdrant), manage cost-efficient GPU scheduling, and collaborate with engineers to enable smooth transitions across workflows.