Principal Researcher – AI Data Center Networking

Principal Researcher – AI Data Center Networking

Role Overview:
Take the lead in shaping the research direction for next-generation AI data center network (DCN) architectures. Focus on developing advanced networking algorithms, protocols, and system-level innovations to support large-scale AI training and inference workloads.

Key Responsibilities:

• Conduct high-impact research on advanced DCN architectures and emerging technologies for large-scale AI systems.

• Collaborate with global research teams, universities, and product development groups to drive technical innovation and align research with real-world implementation.

• Enhance performance and efficiency across parallel computing environments by optimizing the integration of networking, compute, and storage systems.

• Improve inference system performance by incorporating cutting-edge techniques such as prefill-decode disaggregation and KV cache pooling.

• Explore trends in data center networking and evaluate novel communication protocols to accelerate AI training and inference pipelines.

• Research a wide scope of areas including network topologies, traffic management, software-defined networking (SDN), hybrid optical-electrical networks, and in-network computing.

Preferred Qualifications:

• PhD in Computer Science, Electrical/Electronic Engineering, AI, Automation, Mathematics, Physics, or a related field.

• 5+ years of research experience in data center networking or large-scale training/inference systems, either in academia or industry.

• Strong expertise in high-performance interconnects and data center network systems.

• Hands-on experience optimizing parallel computing infrastructure.

• In-depth understanding of large language model (LLM) architectures, parallelism strategies, and model optimization methods.

• Familiarity with distributed deep learning infrastructure and related technologies.

• Strong written and verbal communication skills in English, with the ability to work effectively across global and multicultural teams.

• Proven ability to work independently and contribute within collaborative, cross-functional environments.

Desirable Expertise:

• Broad understanding of system-level challenges in scaling and optimizing large AI training/inference infrastructures.

• Experience with modern networking technologies, including DCN architectures, resource pooling, and optical networking.

• Research background in distributed computing, deep learning systems, high-performance storage, and advanced hardware design.

• Strong foundation in network theory and optimization techniques, including congestion control, traffic scheduling, TCP/RDMA acceleration, and low-latency solutions.

• Familiarity with emerging technologies such as CXL protocol extensions, non-CLOS topologies, GPU/DPU interconnects, and next-gen network hardware.