Senior Software Engineer (AI Inference & Runtime Platform)
Core
Build and operate the AI inference control plane and agent-sandboxing platform, managing GPU workloads, Kubernetes clusters, and secure microVMs for untrusted code execution.
Role type
Senior IC systems engineer (AI inference & runtime platform)
Builds
Inference control plane, agent-sandboxing platform (microVMs), Kubernetes operators, and FastAPI control-plane services
Domain
AI infrastructure, cloud-native systems, security
Deliverable
production ML models
Required skills
Rust, Kubernetes operators/controllers, async runtimes, threat modeling for isolation boundaries, open-weight LLM serving (vLLM/SGLang), distributed systems performance, infrastructure as code (Terraform/OpenTofu)
Preferred skills
Go/C/C++/Zig, Firecracker/Kata/gVisor, secrets management (Vault/KMS), stateful data plane management
Technologies
Rust, Python/FastAPI, Kubernetes, KEDA, Karpenter, AWS/Azure/GCP, OpenTelemetry, pgvector, Neo4j
Responsibilities
Manage serving tier for open-weight models including engine deployment, cold-start strategy, and per-model SLOs; Administer Kubernetes layer for inference and sandbox workloads including operators, autoscaling, and GPU scheduling; Own stateful data plane end-to-end with restore procedures; Oversee sandbox runtime lifecycle and isolation boundary threat model; Direct FastAPI control-plane services and dashboards; Manage open-source posture and release cycles
Seniority
Senior, hands-on IC
