About the role
The team works to make AI inference on the RunPod platform the fastest and most cost-efficient. The engineer measures, diagnoses, and improves LLM serving performance across models and hardware. This work directly impacts customer latency and cost, with a focus on finding and fixing bottlenecks for production systems.
Our summary of the posting; the original is on the RunPod careers page.
What the posting requires
- Required
- vLLM
- Python
- Inference optimization
- Benchmarking
- GPU profiling
- SGLang
- Software engineering
- Quantization
- Performance analysis
- Preferred
- GPU kernels
- Inference
- Multi-node GPU
- CUDA
- ML systems
- High-speed networking
These tags are our reading of the posting. They can miss something; the original posting is the reference.