High Performance Network Engineer AI Inference Network Interface Card
Active AI foundation-model company offering multimodal models, consumer AI products, and an enterprise/developer API platform.
Funding history
About MiniMax Group Inc.
MiniMax develops proprietary multimodal foundation models spanning language, video, speech, and music, and operates AI-native products including MiniMax Agent, MiniMax Design, MiniMax Audio, Talkie, Hailuo AI, and an Open Platform.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will optimize NIC, PCIe, driver, kernel, and RDMA communication performance across single-host and multi-host AI inference clusters. You will improve critical inference communication paths, including KV cache transfer, cross-node collective communication, and model-weight loading. You will investigate packet loss, latency variation, slow nodes, and NIC failures; strengthen monitoring and automated diagnostics; collaborate with inference-framework, hardware, and NIC-vendor partners; and evaluate emerging networking technologies.
Requirements
- 3+ years of relevant experience
- Knowledge of computer networking and operating systems
- Knowledge of RDMA, including QP, Verbs, memory registration, and transport modes
- Experience with RoCEv2 or InfiniBand
- Knowledge of collective communication algorithms including Ring, Tree, Halving-Doubling, and All2All
- Knowledge of Linux networking and the effects of PCIe and NUMA on communication performance
- Ability to independently perform performance tuning and fault troubleshooting
- Proficiency in C or C++ and at least one of Python, Go, or Shell
Responsibilities
- Optimize NIC, PCIe, driver, kernel, and RDMA communication performance
- Improve KV cache transfer, cross-node collective communication, and model-weight loading paths
- Investigate packet loss, latency variation, slow nodes, and NIC failures
- Analyze root causes across NICs, PCIe, drivers, kernels, and applications
- Build monitoring, alerting, automated diagnostics, and performance baselines
- Collaborate with inference-framework, hardware, and NIC-vendor partners
- Evaluate and implement emerging NIC, RDMA, and GPU-interconnect technologies