Inference & Model Serving Jobs
Running models in production — inference engines, model serving, and latency/throughput optimization (vLLM, TensorRT and similar). 45 open now, refreshed daily.
- open roles
- 45
- companies
- 21
- list salary
- 24 · $139K–$560K
- visa mention
- 12
- remote
- 4
Observed across current open postings, refreshed daily — not a survey. Salary band is drawn only from roles that publish a range. Salary breakdown →
Inference and model-serving roles own the production side: getting trained models to answer fast and cheaply under real traffic. That means serving engines and runtimes (vLLM, TensorRT-LLM and the like), continuous batching and KV-cache strategy, quantization, and the latency/throughput trade-offs that decide unit economics for anyone shipping an LLM product. They concentrate at the labs and inference-platform startups whose revenue is literally tokens-per-second — so the work rewards people who reason fluently about both model internals and the systems that run them.
Hiring most for this specialty: Anthropic 9 · Databricks 5 · Together AI 5 · Baseten 4 · CoreWeave 4 · OpenAI 2 · see all who's hiring →