Emerging Network Architect –HPC & AI Infrastructure
Location: Dallas, TX
Employment Type: Direct Hire
Work Arrangement: Hybrid
Overview
Our client is seeking an experienced
Emerging Network Architect to help evaluate, design, and integrate next-generation networking technologies supporting large-scale high-performance computing, AI, and accelerated compute environments.
This role sits at the intersection of
networking, compute, storage, and distributed systems, with a focus on identifying emerging technologies, benchmarking performance, evaluating new architectures, and helping guide long-term infrastructure strategy.
The successful candidate will work closely with internal architects, engineering teams, technology vendors, and customers to translate evolving networking capabilities into scalable, production-ready solutions. This is a highly technical role for someone who enjoys working hands-on with advanced networking technologies while also influencing architecture decisions and vendor roadmaps.
Key Responsibilities
- Evaluate emerging network architectures and technologies that can improve HPC and AI infrastructure performance, including model utilization, throughput, latency, efficiency, and performance per watt.
- Research and assess next-generation networking technologies, interconnects, fabrics, and system architectures.
- Develop strong technical partnerships with leading networking and infrastructure vendors and incorporate future technology roadmaps into platform strategy.
- Recommend and justify hardware and software solutions aligned with performance, scalability, reliability, and efficiency objectives.
- Work with customers and technical stakeholders to understand current and future HPC, AI/ML, scientific computing, and simulation workloads.
- Translate workload requirements into network architecture, topology, bandwidth, latency, and performance requirements.
- Evaluate new networking hardware and software in lab and test environments before broader production deployment.
- Design and execute benchmarks, performance tests, and proof-of-concept environments for emerging technologies.
- Identify network bottlenecks and perform latency, bandwidth, congestion, and performance analysis.
- Partner with compute and storage architects to develop complete end-to-end HPC and accelerated computing architectures.
- Provide technical guidance to infrastructure and engineering teams responsible for deploying selected architectures at scale.
- Continuously evaluate the HPC, AI infrastructure, and networking ecosystem across established vendors, emerging companies, and new technologies.
- Influence vendor product roadmaps through technical feedback, joint evaluations, proof-of-concept testing, and strategic partnerships.
- Support architecture decisions related to high-speed interconnects, cluster fabrics, distributed storage, and GPU-based computing environments.
Qualifications
- Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related technical discipline, or equivalent professional experience.
- Deep expertise in network architectures, high-performance interconnects, and network topologies supporting HPC, AI, or large-scale compute environments.
- Hands-on experience with high-speed networking technologies such as:
- InfiniBand
- NVLink / NVSwitch
- Ethernet / RoCE
- Omni-Path or similar high-performance fabrics
- Strong understanding of packet switching, routing algorithms, flow control, congestion management, and network optimization.
- Experience evaluating the relationship between networking performance and distributed compute workloads.
- Knowledge of performance benchmarking tools and methodologies such as OSU Micro-Benchmarks, MPI benchmarks, STREAM, or similar tools.
- Hands-on experience working in lab, proof-of-concept, or engineering environments testing new or pre-production hardware.
- Strong understanding of distributed and parallel storage architectures and their network requirements, including technologies such as VAST Data, Lustre, or similar high-performance storage platforms.
- Experience designing HPC clusters, accelerated compute platforms, or parallel computing environments.
- Strong Linux knowledge, including system-level optimization, kernel tuning, performance profiling, and troubleshooting.
- Experience analyzing latency, bandwidth, topology, congestion, and overall system performance across large distributed environments.
- Ability to work directly with technical customers and stakeholders to gather requirements and design scalable infrastructure solutions.
Preferred Experience
Experience with emerging technologies such as the following is highly valued:
- CXL
- PCIe Gen 6 and future PCIe architectures
- DPUs / SmartNICs
- GPU networking and scale-out architectures
- NVIDIA networking technologies
- Advanced Ethernet and ultra-low-latency networking
- AI/ML training infrastructure
- Large-scale GPU clusters
- Scientific computing, simulation, or CFD workloads
Ideal Background
The ideal candidate has worked in environments where
network architecture directly impacts compute performance and understands how networking, storage, GPUs, CPUs, and distributed software must work together as a complete system.
Candidates coming from HPC infrastructure, hyperscale computing, AI infrastructure, GPU platforms, semiconductor organizations, networking vendors, research computing, or advanced data center engineering environments may be particularly well aligned.