Member of Technical Staff, Infrastructure Engineer

Odyssey; Zürich; Vor Ort; CHF 120.000 - 170.000; Vollzeit

Zusammenfassung

Odyssey is seeking an engineer to build and operate the compute and infrastructure that powers its world models. You will design scalable, low-latency inference platforms, GPU-based training clusters, and data pipelines, while focusing on performance and reliability.

You will work with researchers and product teams to enable fast experimentation and deployment, emphasizing developer-friendly tooling and efficient resource use.

Qualifikationen

• Proven experience with large-scale distributed GPU workloads.
• Strong Python/Go programming and software engineering practices.
• Hands-on with Docker, Kubernetes and Terraform.
• Collaborative mindset and ability to work with researchers and product teams.

Aufgaben

• Develop and operate a low-latency model inference platform.
• Scale core data processing infra (Flyte, Ray, Kubernetes) for petabyte-scale datasets.
• Build and maintain large-scale GPU-based training clusters for DL, prioritizing usability and reliability.
• Automate infrastructure provisioning, configuration, monitoring, and alerts using IaC.
• Drive performance tuning, cost optimization, and reliability improvements across the stack.
• Collaborate with researchers and product developers to optimize workflows and platform usability.

Kenntnisse

Python

Go

Distributed systems

IaC

Performance tuning

Collaboration

Tools

Docker

Kubernetes

Terraform

Flyte

Ray

Jobbeschreibung

Who We Are

Odyssey is an AI lab pioneering general world models: causal, multimodal systems that learn to predict and interact with the world over long horizons. This foundational technology promises to revolutionize robotics, science, healthcare, education, gaming, defense, and beyond. Odyssey’s founders previously pioneered the most complex application of physical AI: self‑moving cars. They’ve now brought together a world‑class research team from DeepMind, Tesla, Waymo, Meta, Apple, and Wayve, who have made significant contributions to language models (DeepMind Gemini), video models (DeepMind Veo), world models (Wayve GAIA), and autonomous systems (Tesla FSD). Odyssey has raised significant venture capital from GV, Amazon, AMD, EQT, NVIDIA, Natural Capital, In‑Q‑Tel, Elad Gil, Jeff Dean, Guillermo Rauch, Garry Tan, Kyle Vogt, and researchers from OpenAI, DeepMind, MSL, Recursive, and Thinking Machines.

What We’re Looking For

We are looking for an engineer who thrives on building the engines that make groundbreaking research and products possible. You think in systems, love performance, and get energy from turning theoretical bottlenecks into beautifully efficient reality. You’re excited to design and support infrastructure not just for scale, but for speed, creativity, and discovery. You want to build the compute substrate that lets Odyssey’s world models imagine, act, and interact in real time.

What You’ll Do

• Develop and operate our low‑latency model inference platform, ensuring high availability, scalability, and efficient resource utilization for Odyssey’s world models.
• Engineer and scale our core data processing infrastructure (e.g., Flyte, Ray with k8s) to handle petabyte‑scale datasets.
• Design, build, and maintain our large‑scale, GPU‑based training clusters for deep learning, focusing on usability, high throughput and reliability.
• Automate infrastructure provisioning, configuration, monitoring, and alerting using Infrastructure as Code (IaC) principles.
• Drive performance tuning, cost optimization, and reliability improvements across the entire stack.
• Collaborate closely with researchers and product developers to understand their requirements, optimize their workflows, and improve platform usability.

Who You Are

• Motivated by building for the frontier: you want to shape the compute and infrastructure foundation of a lab redefining how people create and interact with media.
• Strong programming skills (e.g., Python, Go, or similar) and a solid understanding of software engineering best practices.
• Deep, hands‑on experience with containerization (e.g., Docker), container orchestration (Kubernetes) and Infrastructure as Code (Terraform).
• Proven experience building and managing large‑scale, distributed systems with GPU computational workloads (e.g., compute platforms, data pipelines, or high‑availability services).
• Experienced in designing infrastructure for ML workloads where performance, parallelism, and data movement are critical.
• A collaborative mindset and excellent communication skills, with a passion for building developer‑friendly platforms.