Platform Engineer

About the role

Platform Engineer

Vectara is the Enterprise Agent Platform that enables businesses to build and deploy governed, grounded, auditable AI agents across SaaS, VPC, and on-prem. We have developed our own models, observability and orchestration layers, guardrails, and context management systems in order to provide our customers with the greatest levels of AI agent accuracy, security, and explainability.

Our founding members include industry veterans and experts in neural information retrieval and distributed systems from Google. We’re a passionate team that’s hyper-focused on solving enterprise-level technology and business problems with AI. Join us!

The Role You'll own the infrastructure that runs our deploy anywhere platform — from Kubernetes clusters serving ML inference at scale to the CI/CD pipelines, IaC, and observability stack that keep it all reliable. This is a hands-on role: you'll write Helm charts and Terraform one day, debug a Kafka consumer lag issue the next, and ship a backend service feature the day after. You'll deploy across AWS, GCP, and on-premises (including air-gapped environments), and you'll participate in an on-call rotation supporting enterprise customers.

What You’ll Do

- Build and maintain infrastructure-as-code (Terraform, Helm) for our AWS EKS and GCP GKE clusters, plus on-premises deployments (including Tanzu and air-gapped environments).

- Own CI/CD pipelines (GitHub Actions, Bazel, ArgoCD) and drive GitOps adoption.

- Deploy, scale, and optimize ML/NLP inference workloads (vLLM, PyTorch, GPU scheduling with various Kubernetes scalers).

- Build and improve observability: Prometheus, Grafana, Datadog,, and OpenTelemetry.

- Collaborate with Field Engineering to support PoCs and platform deployments in customer cloud VPCs and on-prem environments.

- Contribute to backend services (Java 21, Python, gRPC) and platform features.

- Improve system reliability, scalability, and developer experience across the engineering org.

What You’ll Bring (Required):

- 5+ years in platform engineering, DevOps, SRE, or backend infrastructure roles.

- Strong Kubernetes experience (deployment, debugging, scaling — not just `kubectl apply`).

- Hands-on with infrastructure-as-code: Terraform, Helm, or Pulumi.

- Experience with at least one major cloud provider (AWS preferred; GCP or Azure also valued).

- Proficiency in one or more of: Go, Python, Java. Comfortable reading and contributing to backend codebases.

- Working knowledge of CI/CD systems (GitHub Actions, Bazel, ArgoCD, or similar).

- Solid fundamentals in Linux, networking, and distributed systems.

What Sets You Apart (Preferred)

- Experience deploying or operating ML inference workloads (model serving, GPU scheduling, vLLM, TensorFlow Serving, or similar).

- Familiarity with streaming/messaging systems (Kafka, Pulsar) and data stores (MariaDB/PostgreSQL, Aerospike, ClickHouse, OpenSearch).