About the job
Decagon builds agentic AI that resolves customer support conversations end to end, for companies ranging from fast-growing startups to some of the largest financial institutions in the world. Keeping that system fast, reliable, and secure (across our multi-tenant cloud and inside customers' own locked-down environments) is an infrastructure problem, and that's the problem this role owns. You'll build the platforms and abstractions our product teams ship on, and you'll architect and operate the deployments that run our agents inside enterprise customer clouds, where security, compliance, and operational rigor matter as much as speed.
Responsibilities
Design the development and production platforms that power our products, and the abstractions over cloud infrastructure, Kubernetes, and networking that let engineers ship without becoming infrastructure experts. Make sure it all scales to the next order of magnitude as usage grows.
Take end-to-end ownership of deployment architecture in customer-owned cloud environments (VPC configuration, permissioning, networking, provisioning) and the full lifecycle that follows: setup, upgrades, scaling, and incident support. Build the runbooks and automation that make it repeatable.
Treat monitoring, alerting, and rollback as first-class parts of anything you ship, not afterthoughts. Own the reliability of the systems our AI agents depend on in production, where latency, availability, and graceful degradation directly shape the customer experience.
Work directly with customers' platform, security, and DevOps teams to navigate their infrastructure and compliance constraints, and with our Product, Security, Sales, and Customer Success teams to turn customer requirements into concrete deployment plans.
Qualifications
Minimum
4+ years building and operating core infrastructure, platform engineering, or infrastructure/DevOps, ideally with some customer-facing deployment experience.
Deep experience with a major cloud provider (GCP, AWS, or Azure), along with Terraform and Kubernetes at scale.
Strong grasp of cloud networking fundamentals (VPCs, IAM, DNS, load balancing) and how they surface as real deployment constraints.
A track record operating production systems reliably: monitoring, on-call, incident response, and reasoning about failure modes up front.
Comfort navigating ambiguity across a range of stakeholders, from engineers to security and compliance teams, and turning those conversations into actionable plans.
Clear technical writing and a track record of driving adoption across teams.
Comfortable in a fast-moving environment with rapid change.
Preferred
Experience managing deployments in customer-owned cloud environments, including security reviews, compliance requirements, and change management.
Experience building internal platforms or paved roads: service templates, self-serve environments, CI/CD pipeline design, deployment automation.
Familiarity with observability and incident management in distributed systems (Prometheus, Grafana, Datadog, or similar).
Infrastructure-as-code with a security-minded approach to supply chain (provenance, secrets, least privilege).
Experience operating latency-sensitive or ML/AI-serving workloads in production.
Experience using AI-assisted tooling to make yourself and your team dramatically more effective.