AI Infrastructure Engineer [Tooling]

The Role

As an AI Infrastructure Engineer, you'll design, build, and own the tooling and services that take Tessera's AI agents from a developer's laptop to production in customer clouds. You'll write real software in Python or Go - CLIs, controllers, and internal services - and run it on Kubernetes alongside LLM serving stacks like vLLM.

This is a hands-on IC role open at mid to senior level. Mid-level engineers will own well-scoped components end to end; senior engineers will also set technical direction, make architectural calls across the platform, and mentor others. We'll calibrate level based on experience and interviews.

What You'll Do

AI infrastructure tooling

• Build and maintain internal CLIs, SDKs, and services in Python or Go that let engineers provision, deploy, test, and debug AI agents across environments
• Deploy, scale, and tune LLM inference using vLLM (and similar serving stacks), including GPU scheduling, autoscaling, batching, and cost/latency trade-offs
• Build infrastructure for multi-agent systems: agent runtimes, tool and connector execution, sandboxing, state management, and safe rollout of new agent versions
• Own MLOps workflows: model and prompt versioning, evaluation pipelines, artifact registries, and promotion from staging to production
• Write Kubernetes operators and custom controllers (CRDs) that encode how Tessera's AI workloads are deployed and managed

Cloud platform and reliability

• Design, build, and operate production-grade AI agent infrastructure on Kubernetes and Terraform
• Support Tessera's portable infrastructure across Azure, Oracle Cloud, GCP, and AWS, with primary ownership of at least one cloud variant
• Improve observability for AI systems: tracing of agent and LLM calls, token and GPU utilization, latency, error rates, and cost
• Strengthen security and reliability for workloads running in customer environments under strict compliance constraints

Automation and developer experience

• Design and own CI/CD for infrastructure, services, and model/agent releases, using GitOps and pipeline-as-code
• Build self-service tooling that removes manual work: environment provisioning, secret rotation, dependency updates, compliance checks, and release management
• Partner closely with AI, product, and platform engineers to unblock delivery
• Champion an automation-first mindset: if something is done twice, it should be automated; if it's automated, it should be observable and maintained like production code

Must-have Qualifications

• 5+ years of experience in infrastructure, platform, or backend engineering (7+ years for senior)
• AI infrastructure tooling: strong software engineering in Python or Go, with a track record of building and shipping tools, CLIs, or services that other engineers rely on
• Hands-on experience running LLM workloads in production, including serving with vLLM or a comparable stack (TGI, Triton, SGLang, Ray Serve)
• Experience building infrastructure for AI agents or LLM applications: orchestration, tool calling, evaluation, or agent runtimes
• Practical MLOps experience: model/prompt versioning, evaluation and release pipelines, and observability for ML systems
• Strong hands-on experience with Kubernetes, including writing or extending operators/controllers (Kubebuilder, Operator SDK, controller-runtime, or Kopf)
• Strong hands-on experience with Terraform and at least one of Azure, Oracle Cloud, GCP, or AWS
• Proven experience designing and owning CI/CD systems, GitOps-driven deployments, and release automation
• Solid networking fundamentals
• Proven experience deploying and operating real production systems

Nice-to-have Qualifications

• GPU infrastructure experience: NVIDIA GPU Operator, MIG, KServe, or Karpenter-style autoscaling for inference
• Familiarity with agent frameworks and protocols (LangGraph, Model Context Protocol, OpenAI Agents SDK, or similar)
• LLM observability tooling such as OpenTelemetry, Langfuse, or Arize
• Experience with FluxCD, ArgoCD, or other GitOps controllers
• Cloud security and DevSecOps: security gates, SBOM generation, and compliance checks in CI/CD
• Deep networking expertise (service mesh, private connectivity, multi-cloud networking)
• Open-source contributions to AI infrastructure or Kubernetes ecosystem projects
• A passion for clean, unified infrastructure and a builder mindset

Who You Are

• A builder who writes production-quality code, not just configuration
• Curious about the fast-moving AI infrastructure landscape and eager to learn new tools
• Comfortable owning and defending infrastructure decisions, with clear reasoning about trade-offs
• Experienced in startups and/or big tech environments
• Highly self-motivated with a strong ownership mindset
• Someone who enjoys building foundational systems that other engineers rely on