网络架构师负责人（AI SuperNIC） – 欧洲 / 以色列 / 美国（可能远程）
Lead Networking Architect (AI SuperNIC) – Europe / Israel / US (Remote Possible)
我们的客户是一家快速发展的深度科技公司，正在构建下一代人工智能基础设施，现正寻找一位卓越的网络架构负责人，以推动新型 AI SuperNIC 和扩展性网络解决方案的架构设计。
这是公司最重要的职位，也是难得的机会，可以影响下一代数据中心中使用的尖端人工智能加速平台的架构。
公司简介
我们的客户是一家成立5年的 AI 半导体初创公司，专注于开发：
- AI SuperNIC
- AI 计算解决方案
- 下一代 CPU 和网络技术
公司已完成成功的流片并推出了产品。
全球约70名员工；一个高度技术化、资深且注重创新的环境。
扁平化管理，决策迅速，与 VLSI 副总裁和 CEO 直接合作。
职位概述
作为网络架构负责人，你将定义并塑造公司下一代 AI SuperNIC 的网络架构——专注于超低延迟、扩展性通信和先进的数据中心网络协议。
你将与 VLSI、固件、软件和 AI 平台团队紧密合作，并在关键架构决策中发挥核心作用。
主要职责
- 设计 AI SuperNIC 的扩展性通信层。
- 定义硬件、固件和软件的系统级网络需求。
- 领导性能建模、仿真、基准测试和优化。
- 评估并集成高级标准：RoCEv2、RDMA、UEC、UALink、NVLink、NVMe-oF。
- 与内部研发团队及外部合作伙伴/客户协作。
- 推动数据中心互连和人工智能网络解决方案的创新。
要求
- 10年以上网络、系统架构、NIC/DPU/SoC 开发或高性能计算（HPC）网络经验。
- 精通 RDMA、RoCEv2、NVMe-oF 和高性能网络。
- 深入理解网络协议、扩展性架构和低延迟设计。
- 具有跨硬件+固件+驱动/软件的经验。
- 能够领导架构定义、系统建模和性能分析。
- 具有 NIC、SmartNIC、DPU、网络芯片或数据中心互连经验者优先。
这个职位为何特别
- 参与具有巨大行业影响力的最前沿人工智能网络技术。
- 直接参与架构决策——无官僚主义，执行速度快。
- 开发一款挑战行业的产品。
查看英文原文
Our client, a fast-growing deep-tech company building next-generation AI infrastructure, is looking for an exceptional Lead Networking Architect to drive the architecture of a new AI SuperNIC and scale-out networking solution.
This is the highest-priority role in the company and a rare opportunity to influence the architecture of a cutting-edge AI accelerator platform used in next-generation data centers.
About the Company
Our client is a 5-year-old AI semiconductor startup developing:
- AI SuperNIC
- AI compute solutions
- Next-generation CPU and networking technologies
The company has already completed a successful tape-out and launched a working product.
~70 employees globally; a highly technical, senior, and innovation-driven environment.
Flat hierarchy, fast decision-making, and direct collaboration with VP VLSI and the CEO.
Role Overview
As the Lead Networking Architect, you will define and shape the networking architecture of the company’s next-generation AI SuperNIC - focused on ultra-low latency, scale-out communication, and advanced data-center networking protocols.
You will collaborate closely with VLSI, firmware, software, and AI platform teams, and play a critical role in key architectural decisions.
Key Responsibilities
- Architect the scale-out communication layer of the AI SuperNIC.
- Define system-level networking requirements for hardware, firmware, and software.
- Lead performance modeling, simulations, benchmarking, and optimization.
- Evaluate and integrate advanced standards: RoCEv2, RDMA, UEC, UALink, NVLink, NVMe-oF.
- Collaborate with internal R&D teams and external partners/customers.
- Drive innovation in data-center interconnects and AI networking solutions.
Requirements
- 10+ years in networking, system architecture, NIC/DPU/SoC development, or HPC networking.
- Strong expertise in RDMA, RoCEv2, NVMe-oF, and high-performance networking.
- Deep understanding of networking protocols, scale-out architectures, and low-latency design.
- Experience working across hardware + firmware + driver/software.
- Ability to lead architectural definitions, system modeling, and performance analysis.
- Experience in NIC, SmartNIC, DPU, networking silicon, or data-center interconnects is required.
Why This Role Is Exceptional
- Work on state-of-the-art AI networking technology with massive industry impact.
- Direct influence on architectural decisions — no bureaucracy, fast execution.
- Build a product that challenges industry leaders (Nvidia, Marvell, Intel).
- Join a highly senior engineering team where your expertise truly matters.
- Opportunity to shape a new category of AI infrastructure.
Location & Work Model
- Israel – onsite
- Europe / US – remote acceptable
- Must be able to work Israel-aligned working hours (flexible)
B2B arrangements in Europe; direct employment in Israel; other geographies flexible.
Compensation
The company is ready to offer a highly competitive package for the right expert.
Compensation is flexible and depends on experience and impact level.
Hiring Process:
- 30-minute prescreen call with an external recruiter – evaluation of core networking and architecture experience.
- Initial phone interview with HR to confirm background and role expectations.
- Technical interview loop with senior architects and engineering leadership.
- Final interview with the COO (mandatory step for all architect-level roles).
- HR interview (if needed) + offer.
Interested?
If you are a senior networking architect passionate about building the future of AI infrastructure, we’d love to hear from you.
Originally posted on Himalayas