Software Engineer - Data Infra Reliability

Luma AI is dedicated to building multimodal AI to expand human imagination and capabilities, with a focus on vision-based foundation models. This role seeks a Data Reliability Engineer to ensure the resilience, automation, and scalability of petabyte-scale data pipelines. The engineer will apply Site Reliability Engineering (SRE) principles to create self-healing data systems, enabling researchers to train on massive datasets without interruption, thereby hardening the backbone of Luma’s intelligence.

Must Have

• Automate data infrastructure using Infrastructure-as-Code (IaC) principles with Terraform.
• Build reliability and fault tolerance into core data ingestion and processing workflows.
• Operate and optimize large-scale Kubernetes clusters and Ray deployments.
• Establish Service Level Objectives (SLOs) and observability standards (Prometheus/Grafana).
• Diagnose root causes of complex infrastructure failures in distributed storage and compute systems.
• Deep SRE/DevOps proficiency, including Linux, networking, and automation.
• Extensive experience with Terraform, Ansible, or similar tools to manage cloud environments (AWS/GCP).
• Managed Kubernetes in production and understand its internals.
• Proficiency in writing high-quality Python code for automation, tooling, and infrastructure management.
• Understanding of specific challenges of stateful data systems and high-throughput storage (S3/Object Store).

Good to Have

• Experience managing GPU clusters or AI/ML workloads.
• Background in both Software Engineering and Operations (DevOps).
• Experience with high-performance networking (InfiniBand/RDMA).

About Luma AI

Luma's mission is to build multimodal AI to expand human imagination and capabilities. We believe that multimodality is critical for intelligence. To go beyond language models and build more aware, capable, and useful systems, the next step function change will come from vision. So, we are working on training and scaling up multimodal foundation models for systems that can see and understand, show and explain, and eventually interact with our world to effect change.

Where You Come In

As our models scale to "omni" capabilities, our data infrastructure must be unbreakable. We are looking for a Data Reliability Engineer who brings a Site Reliability Engineering (SRE) mindset to the world of massive-scale data. You will be responsible for the resilience, automation, and scalability of the petabyte-scale pipelines that feed our research. This is not just about keeping the lights on; it’s about treating infrastructure as code and building self-healing data systems that allow our researchers to train on massive datasets without interruption. Whether you are a junior engineer with a passion for automation or a seasoned SRE veteran, you will play a critical role in hardening the backbone of Luma’s intelligence.

What You'll Do

• Automate Everything: Apply Infrastructure-as-Code (IaC) principles using Terraform to provision, manage, and scale our data infrastructure.
• Harden Data Pipelines: Build reliability and fault tolerance into our core data ingestion and processing workflows, ensuring high availability for research jobs.
• Scale Kubernetes & Ray: Operate and optimize large-scale Kubernetes clusters and Ray deployments to handle bursty, high-throughput workloads.
• Define Reliability: Establish Service Level Objectives (SLOs) and observability standards (Prometheus/Grafana) for our data platforms.
• Debug & Heal: serve as the first line of defense for complex infrastructure failures, diagnosing root causes in distributed storage and compute systems.

Who You Are

• Deep SRE/DevOps proficiency: You live and breathe Linux, networking, and automation.
• Infrastructure-as-Code Native: You have extensive experience with Terraform, Ansible, or similar tools to manage complex cloud environments (AWS/GCP).
• Kubernetes Expert: You have managed Kubernetes in production and understand its internals, not just how to deploy containers.
• Python Proficiency: You can write high-quality Python code for automation, tooling, and infrastructure management.
• Data-Minded: You understand the specific challenges of stateful data systems and high-throughput storage (S3/Object Store).

What Sets You Apart (Bonus Points)

• Experience managing GPU clusters or AI/ML workloads.
• Background in both Software Engineering and Operations (DevOps).
• Experience with high-performance networking (InfiniBand/RDMA).

Compensation

The base pay range for this role is $220,000 – $280,000 per year.