Salesforce
Principal Software Engineer, Site Reliability (Agentforce Ops)
$197K–$345K
California - San Francisco Metro - Remote +6 · Office Tech-Flexible · Principal · Full-time · Posted Oct 29, 2026
- Location
- California - San Francisco Metro - Remote
Virginia - Washington DC Metro - Remote
Washington D.C. - Remote
Texas - Austin Metro - Remote
Washington - Seattle
New York - New York City Metro - Remote
Massachusetts - Boston Metro - Remote - Experience
- 9+ years; SRE, Production Engineering, or Backend Engineering.
- Education
- M.S. in Computer Science or equivalent practical experience is preferred.
About the job
Leads reliability, scaling, performance, and infrastructure automation for the Agentforce Operations platform and its AI/ML systems.
What you'll do
- Own reliability roadmaps and guide technical direction for scaling product areas to global, enterprise-grade availability.
- Improve infrastructure automation, observability, and production resilience; build tools and telemetry to prevent and diagnose incidents.
- Tune SQL, API latency, and cross-service performance, and support AI/ML infrastructure, GPU resources, and data pipelines.
What they ask for
- 9+ years of experience in SRE, Production Engineering, or Backend Engineering focused on operations and scale.
- Production-grade coding expertise in Golang, TypeScript, or Python.
- Strong proficiency in Kubernetes, Terraform/OpenTofu, and AWS/GCP/Azure.
- Experience with SLO-based monitoring and alerting, distributed systems, and using modern AI development tools.
Nice to have
- M.S. in Computer Science or equivalent practical experience; PostgreSQL at scale.
- Experience with Temporal, Istio, Prometheus, Grafana, Pagerduty, Elastisearch or Typesense, or Salesforce infrastructure, Hyperforce, or Data Cloud.
Pay
$197,300 - $313,700 annually (base salary); $237,700 - $344,700 annually (base salary)
Tools
Golang, TypeScript, Python, Kubernetes, Terraform/OpenTofu, AWS/GCP/Azure, PostgreSQL, Temporal
AI focus
Supports scaling AI/ML infrastructure, including GPU resources and data pipelines, and uses AI development tools such as Claude Code and Codex to automate operations and accelerate infrastructure delivery.
Model-written summary · the original posting is authoritative