About the role
The Inference Cloud Engineering team develops the technology for d-Matrix's inference services. The Inference Performance Engineer diagnoses and resolves performance issues across the entire serving path, ensuring workloads perform optimally on the platform. This involves profiling, hypothesizing, testing, and collaborating with other teams to address root causes.
Our summary of the posting; the original is on the d-Matrix careers page.
What the posting requires
- Required
- Linux internals
- Kubernetes
- Python
- AI tools
- NUMA
- Performance tuning
- C++
- Agentic tools
- CPU pinning
- Preferred
- vLLM
- KV-cache management
- GPU programming
- Rust
- SGLang
- Disaggregated serving
These tags are our reading of the posting. They can miss something; the original posting is the reference.