Staff Site Reliability Engineer | Infrastructure and Observability

The third Harvey role in today's list, this one on site reliability. The tooling is named specifically rather than generically, which makes it easy to judge fit: infrastructure as code with Pulumi, Terraform or CloudFormation, observability with Datadog and Sentry, incident response with PagerDuty and Incident.io, cloud across Azure, GCP or AWS, and programming in Python, Bash or Go. Beyond tools it asks for a proven track record of diagnosing complex system problems and implementing durable solutions, which is the part that actually distinguishes senior SRE work from operations. Stated bar is twelve or more years in SRE or similar roles supporting production environments, with proven ability to mentor and guide technical teams. Good fit if you want reliability ownership on an AI product at genuine seniority.

Published requirements: 12+ years of experience in site reliability engineering or similar roles supporting production environments, with proven ability to mentor and guide technical teams; expertise in infrastructure as code tools including Pulumi, Terraform and CloudFormation; deep familiarity with observability tooling including Datadog and Sentry, and with incident response practice including PagerDuty and Incident.io; proficiency with cloud platforms across Azure, GCP and AWS; strong programming skills in Python, Bash, Go or similar; a proven track record of diagnosing complex system problems and implementing durable solutions; and a solid understanding of CI/CD, Kubernetes, containerisation, networking, databases and cloud security principles. Strong problem solving and meticulous attention to detail are also stated. Location is Bengaluru. The mentoring requirement appears in the experience line itself rather than as a separate management duty, and no reporting line, hiring responsibility or performance review duty is stated anywhere, so we have published this as a senior individual contributor role. Confirm if that distinction matters to you. What is genuinely useful about this posting: naming the actual tools removes the usual guesswork. If your background is AWS and Terraform and this team runs Azure and Pulumi, you can judge the gap yourself rather than discovering it in a screen. The phrase durable solutions is the one to prepare for in interview, since it signals they will probe how you prevented a class of incident from recurring rather than how quickly you restored service. Employer context: Harvey builds AI tools for legal work, and all three of its engineering roles today sit at 10+ or 12+ years, so this is a small senior team rather than a large mixed one. Expect broad ownership and real on call. Good fit for an SRE with twelve or more years who wants reliability ownership rather than a ticket queue.