Tesla Cloud Platform (TCP) is Teslas internal cloud that powers compute, storage, and networking at global scale across multiple datacenters. We are building the next generation of our hypervisor platform on top of KVM to replace legacy virtualization and deliver a purpose-built, high-performance compute layer for Teslas workloads.

Responsibilities

Design and build Teslas next-generation hypervisor platform on top of KVM, including VM lifecycle management, resource scheduling, live migration, and host management
Develop and optimize low-level virtualization components including QEMU/KVM integration, virtio device management, and guest-host interfaces
Build orchestration services and APIs (Go) that power VM provisioning, placement, networking, and storage attachment across Teslas global datacenter fleet
Contribute to the architecture of TCPs compute layer to meet strict requirements for performance, multi-tenancy, and scalability across global datacenters
Integrate observability and monitoring directly into the hypervisor and platform services to provide visibility for both customers and operators
Partner with networking, storage, and infrastructure teams to deliver capabilities such as SR-IOV, custom images, vGPU passthrough, and live migration

3+ years of professional experience in systems-level software development with C, C++, Go (Golang), or Rust, including low-level systems and networking programming
Hands-on experience with KVM, QEMU, libvirt, or similar hypervisor technologies. Contributions to upstream KVM, QEMU, or Linux kernel virtualization subsystems are a strong plus
Understanding of Linux kernel internals relevant to virtualization: namespaces, cgroups, memory management, device passthrough (VFIO), and virtio
Experience with VM lifecycle management, live migration, resource scheduling, or building control planes for virtualization platforms
Solid understanding of networking concepts relevant to virtualization: bridging, VLAN, SR-IOV, Open vSwitch, or DPDK
Knowledge of distributed systems, Infrastructure as Code, and observability in large-scale environments
Proven ability to design and build highly scalable and reliable systems