You will optimize the inference stack to improve throughput and reduce latency for our AI models. This role involves hands-on work with real ML systems and requires a strong technical background in infrastructure.
Details
-
Location: Austria, France, Germany, Italy, Netherlands, Poland, Spain, United Kingdom
-
Salary: Competitive, depending on experience
-
Experience: 5+ years in ML systems
-
Skills: Python, Kubernetes, PyTorch, vLLM, SGLang, TensorRT-LLM
The work
-
Optimize throughput through continuous batching and scheduler tuning
-
Reduce latency by profiling and addressing bottlenecks
-
Enhance KV cache performance with advanced caching strategies
-
Minimize cold start times for large checkpoints