AI Infrastructure Engineer Jobs (2026)

This hub collects listings that center on the compute layer under AI: GPU clusters, distributed training, inference serving, ML platforms, and the networking and storage that feed them. It overlaps with the MLOps hub, since about 370 of the roughly 470 listings here are also categorized as MLOps or ML infrastructure, but the emphasis leans toward the hardware and platform end, where a senior hardware engineer for GPU and AI infrastructure sits next to an AI platform engineer.

As of September 2026 there are about 470 live listings and roughly 50 were posted in the last week. Fifty-four percent are tagged remote. The US names about 200 of the listings, with the UK, India, Germany and France each in the 10 to 20 range. SentinelLabs is the top poster with about 16 roles, then SpaceX, Anduril, Nuro, Highmetric, Nebius and TRM Labs, so security, aerospace, defense, autonomy and cloud infrastructure companies all appear near the top.

Mid level and senior roles are tied at about 140 each, staff and principal roles number about 100, and lead or manager titles about 70. Recent postings include full stack engineer on an ML platform, senior or principal infrastructure engineer for an ML platform, and senior software engineer in ML infrastructure. About 17 percent of listings state a USD salary, with a median midpoint near $202,500 a year among those; the rest do not publish a figure.

Search this hub with the level filter, and use the keyword box for the specific layer you work in, such as Kubernetes or CUDA. The Kubernetes ML jobs and CUDA engineer jobs hubs are narrower cuts of the same territory, and the MLOps hub is the broader one. If remote matters, the remote MLOps hub keeps only the remote tagged roles.