Senior SRE - AI-Native Infra & Observability

Formation Bio; San Francisco (CA); Hybrid; USD 186,000 - 232,000; Full time

Job summary

Formation Bio is seeking a Senior Site Reliability Engineer to build and operate a scalable, secure, and observable infrastructure platform across product apps, data systems, and ML workloads. You will collaborate with Product, Data Engineering, and Data Science to design resilient systems and leverage AI tooling to accelerate development while ensuring reliability.

The role emphasizes cloud-first architecture, IaC, CI/CD, and strong incident response with a bias for automation.

Qualifications

• 5+ years of relevant experience in Site Reliability Engineering, infrastructure, systems, DevOps, or a similar discipline.
• Production experience operating cloud infrastructure and distributed systems, with strong operational and reliability judgment.
• Experience with advanced diagnostics, incident response, root cause analysis, observability, and automation.
• Experience with AWS and Snowflake. Experience with Azure, GCP, and/or Vercel is a plus.
• Working experience with Docker, GitHub, Kubernetes, Python, Terraform or OpenTofu, and virtual networking. Terragrunt is a plus
• Experience supporting production ML or AI workloads, MLOps infrastructure, workflow orchestration, model serving, or a related platform is a plus. Strong SRE and infrastructure fundamentals are the core requirements.
• Experience managing shared COTS and FOSS software applications in addition to our internal software and tools
• Daily fluency with AI tools, including LLMs and agentic coding systems, paired with strong engineering judgment and a high bar for validation.
• Exceptional collaboration and communication skills across engineering, Data Science, Data Engineering, Security, and non-technical partners.
• Experience operating infrastructure in a regulated or validated environment is a plus.

Responsibilities

• Own the infrastructure and operational platform for shared engineering workloads, including compute, runtime environments, orchestration, deployment, observability, access controls, and reliability.
• Build and operate secure, observable, reliable infrastructure for product applications, containerized services, internal tools, data systems, ML pipelines, inference, and agentic software.
• Research, develop, and maintain core AWS infrastructure and additional cloud outposts for services across development, staging, and production.
• Create, review, maintain, and optimize infrastructure as code, CI/CD pipelines, and reusable platform patterns.
• Establish strong operational practices, including SLOs, monitoring, alerting, runbooks, incident response, root cause analysis, and post-incident follow-through.
• Work with Product engineering, Data Engineering, and Data Science to evaluate, negotiate, and implement architecture and infrastructure for product software, data systems, model training, and inference.
• Use AI tools to accelerate infrastructure development, investigate incidents, improve documentation, build automation, and make operational improvements while validating their output.
• Participate in the support rotation and incident response. Maintain automation bias while knowing when ClickOps is appropriate.
• Write and review requirements, design documents, and operating procedures. Share knowledge and mentor engineers on infrastructure and SRE fundamentals.

Skills

5+ years in SRE

Cloud infrastructure

Observability

Automation

Cross-team collaboration

Tools

Docker

GitHub

Kubernetes

Terraform/OpenTofu

Terragrunt

AWS

Snowflake

Azure

GCP

Vercel

Job description

Formation Bio is seeking a Senior Site Reliability Engineer to build and operate a scalable, secure, and observable infrastructure platform across product apps, data systems, and ML workloads. You will collaborate with Product, Data Engineering, and Data Science to design resilient systems and leverage AI tooling to accelerate development while ensuring reliability.

The role emphasizes cloud-first architecture, IaC, CI/CD, and strong incident response with a bias for automation.