Who's hiring GPU, inference & ML-systems engineers right now
31 AI labs and infrastructure companies have 142 open GPU, CUDA, ML-systems, inference and performance engineering roles between them. Ranked by openings, refreshed daily from each company's own public job feed.
- open roles
- 142
- companies
- 31
- list salary
- 75 · $120K–$850K
- visa mention
- 34
- remote
- 12
Observed across current open postings, refreshed daily — not a survey. Salary band is drawn only from roles that publish a range. Salary breakdown →
The useful question for someone in this field usually isn't “what jobs exist” — a search box answers that — but which companies are actually staffing up, and for what. That is what this page tracks. The shape of the demand is fairly consistent: dedicated AI-silicon companies (the Graphcores and Tenstorrents of the list) skew heavily toward performance, compiler and kernel work, because their whole product is making specific hardware fast. GPU-cloud and infrastructure providers (CoreWeave, Nebius and similar) hire for the systems and reliability side — operating accelerator fleets, not designing them. Frontier labs concentrate in inference and ML-systems, where the bottleneck is serving large models efficiently and training them across thousands of GPUs.
Two numbers worth reading off the table: how many of a company's roles disclose a salary band (a rough proxy for hiring maturity and US-comp transparency norms) and how many mention visa sponsorship or relocation — the single most decision-relevant fact for an engineer hiring across borders. Counts move as companies open and close postings; this is a snapshot of the current open set, not a survey.
| Company | Open | GPU | ML-sys | Inf | Perf | Rem | Visa | $ |
|---|---|---|---|---|---|---|---|---|
| Anthropic | 17 | 2 | 1 | 9 | 7 | 1 | 17 | 14 |
| Graphcore | 14 | · | · | · | 13 | · | 5 | · |
| Tenstorrent | 14 | · | · | 1 | 11 | · | · | 9 |
| CoreWeave | 10 | 3 | · | 4 | 4 | · | · | 9 |
| Together AI | 9 | 3 | · | 5 | 1 | · | · | 7 |
| Nebius | 8 | 4 | 1 | 1 | 1 | 7 | · | 3 |
| OpenAI | 8 | 1 | 2 | 2 | 4 | · | 3 | 1 |
| d-Matrix | 7 | · | · | 1 | 4 | · | · | · |
| Databricks | 7 | · | · | 5 | 2 | · | · | 7 |
| Baseten | 5 | 1 | · | 4 | · | · | · | · |
| MatX | 5 | · | · | · | 4 | · | · | 5 |
| SambaNova | 5 | · | 2 | · | 4 | 1 | · | 4 |
| xAI | 4 | 1 | 1 | 2 | 1 | · | · | 3 |
| Prime Intellect | 3 | 1 | · | 1 | · | 1 | 2 | · |
| Scale AI | 3 | · | 2 | 1 | · | · | · | 3 |
| Cohere | 2 | · | · | 1 | 1 | · | · | · |
| Crusoe | 2 | 1 | · | · | · | · | · | 2 |
| EnCharge AI | 2 | · | · | 1 | 1 | 2 | · | 2 |
| Etched | 2 | · | · | · | 1 | · | 2 | · |
| Fireworks AI | 2 | · | · | · | 1 | · | · | 2 |
| FriendliAI | 2 | 1 | · | 1 | 1 | · | · | · |
| Thinking Machines | 2 | · | · | 1 | 1 | · | 2 | 2 |
| Anyscale | 1 | · | · | 1 | · | · | 1 | · |
| Black Forest Labs | 1 | · | · | 1 | · | · | · | 1 |
| FluidStack | 1 | · | · | · | · | · | · | · |
| Lightmatter | 1 | · | 1 | · | · | · | · | · |
| Lightning AI | 1 | 1 | · | · | · | · | 1 | 1 |
| Modal | 1 | · | · | 1 | · | · | · | · |
| Perplexity | 1 | · | · | 1 | · | · | · | · |
| SF Compute | 1 | 1 | · | · | · | · | 1 | · |
| World Labs | 1 | 1 | · | 1 | 1 | · | · | · |
Browse by cut: Salaries · Remote · Europe · Senior · Staff · GPU & CUDA Engineering · ML Systems & Infrastructure · Inference & Model Serving · Performance & Kernel Engineering