Senior Solutions Architect – AI Infrastructure Security (remote in Europe)

免费注册后解锁"你的匹配分 vs 高分画像"与个人技能缺口。

About the role We are seeking a Senior Solutions Architect with deep security expertise to join the Voyager team in the Mirantis Office of the CTO. You will own the security solutioning of k0rdent AI, our platform for building and operating GPU clouds and AI factories, from metal to model. You will define how k0rdent AI environments are secured end to end: the hardware and firmware they boot from, the hosts, virtual machines and Kubernetes clusters they run, the networks and storage they share, the secrets and certificates that hold them together, and the AI models and applications they serve. You will turn that design into published reference architectures, working code and proofs of concept that run on real hardware with our customers and partners. As part of the Office of the CTO, research is a core part of the job. You will explore threats, designs and technologies before customers ask for them, prototype ideas that may not ship, and challenge established practice when a better approach exists. Your work will shape where k0rdent AI goes next, not only how it is deployed today. This role combines research with hands-on, outward-facing work. You will split your time between exploring new designs, writing architecture, building and validating it in the lab, and explaining it to engineers, security teams, executives and conference audiences. You will work with Solutions Architects, Partner Management and Engineering colleagues across many countries and time zones. Key Responsibilities Reference architecture Design and publish security reference architectures and solution designs for k0rdent AI, covering single-tenant private AI clouds, multi-tenant GPU clouds and hybrid deployments that span public cloud and on-premises infrastructure. Define a defence-in-depth model across the stack, from hardware root of trust and firmware to hosts, virtual machines, Kubernetes, networks, storage, models and applications. Define tenant isolation boundaries and their guarantees, and document where each boundary is strong, where it is weak, and what it costs to strengthen. Map designs to the controls and frameworks customers must meet, and document trade-offs (risk, cost, performance, operability) with defensible reasoning a customer can follow. Platform and infrastructure security Bare-metal and host security: secure and measured boot, TPM-based attestation, BMC and Redfish hardening, firmware integrity, and hardened Linux hosts. Virtual machine security: hypervisor and KubeVirt hardening, VM isolation, and the security implications of PCI passthrough and SR-IOV for GPUs and NICs. Kubernetes security: RBAC, Pod Security Standards, admission control and policy-as-code, runtime security, and secure multi-cluster and multi-tenant operation. Secrets management: secret stores, KMS and HSM integration, envelope encryption, secret delivery to workloads, and rotation. Certificates and PKI: internal certificate authorities, automated issuance and rotation, mTLS between services, and workload identity. Network security: segmentation and zero-trust designs, Kubernetes network policy, tenant isolation on front-end and GPU fabrics, and DPU-based security enforcement. Storage security: encryption at rest and in transit, key management, and tenant isolation on shared block, file and object storage used for training data and model weights. Public cloud security: secure landing zones, identity and access management, and network controls on major public clouds, and how they extend to hybrid deployments. AI model and application security Define how Transformer-based models are protected through their lifecycle: provenance and signing of model weights, integrity of training and fine-tuning data, and secure model registries. Secure inference services and AI applications against threats such as prompt injection, data and model exfiltration, model theft and abuse of agentic tool access. Design isolation for shared GPU infrastructure (MIG, vGPU, passthrough) and ass

关于该职位 我们正在寻找一位具备深厚安全专业知识的高级解决方案架构师，加入Mirantis CTO办公室的Voyager团队。你将负责k0rdent AI的安全解决方案设计——这是我们从硬件到模型构建和运营GPU云与AI工厂的平台。你将定义k0rdent AI环境如何实现端到端安全：从它们启动所依赖的硬件和固件，到它们运行的主机、虚拟机和Kubernetes集群，再到它们共享的网络和存储、将它们维系在一起的密钥和证书，以及它们所服务的AI模型和应用。你将把这些设计转化为已发布的参考架构、可运行的代码和概念验证，并在真实硬件上与我们的客户和合作伙伴一起运行。作为CTO办公室的一部分，研究是这项工作的核心。你将在客户提出需求之前探索威胁、设计和技术，对可能不会发布的想法进行原型验证，并在存在更好方法时挑战既定实践。你的工作将塑造k0rdent AI的未来方向，而不仅仅是它今天的部署方式。这个角色将研究与动手实践、面向外部的工作相结合。你将在探索新设计、撰写架构文档、在实验室中构建和验证，以及向工程师、安全团队、高管和会议听众进行讲解之间分配时间。你将与遍布多个国家和时区的解决方案架构师、合作伙伴管理及工程同事合作。 关键职责 参考架构 为k0rdent AI设计并发布安全参考架构和解决方案设计，涵盖单租户私有AI云、多租户GPU云以及跨公共云和本地基础设施的混合部署。定义跨越整个技术栈的纵深防御模型，从硬件信任根和固件到主机、虚拟机、Kubernetes、网络、存储、模型和应用。定义租户隔离边界及其保证，并记录每个边界在何处强、何处弱，以及加强它的成本。将设计映射到客户必须满足的控制措施和框架，并记录权衡（风险、成本、性能、可运维性），提供客户能够理解的、有说服力的推理。 平台与基础设施安全 裸机和主机安全：安全与度量启动、基于TPM的证明、BMC和Redfish加固、固件完整性，以及加固的Linux主机。虚拟机安全：hypervisor和KubeVirt加固、VM隔离，以及GPU和NIC的PCI直通与SR-IOV的安全影响。Kubernetes安全：RBAC、Pod安全标准、准入控制和策略即代码、运行时安全，以及安全的多集群和多租户运营。密钥管理：密钥存储、KMS和HSM集成、信封加密、向工作负载分发密钥，以及轮换。证书与PKI：内部证书颁发机构、自动化签发与轮换、服务间mTLS，以及工作负载身份。网络安全：分段与零信任设计、Kubernetes网络策略、前端和GPU网络上的租户隔离，以及基于DPU的安全执行。存储安全：静态和传输中加密、密钥管理，以及用于训练数据和模型权重的共享块、文件和对象存储上的租户隔离。公共云安全：主要公共云上的安全着陆区、身份与访问管理、网络控制，以及它们如何扩展到混合部署。 AI模型与应用安全 定义基于Transformer的模型在其生命周期中如何受到保护：模型权重的来源与签名、训练和微调数据的完整性，以及安全的模型注册表。保护推理服务和AI应用免受提示注入、数据和模型窃取、模型盗用以及代理工具访问滥用等威胁。为共享GPU基础设施（MIG、vGPU、直通）设计隔离，并 ass

⚑ 岗位信息收录自公开招聘页（Himalayas），以雇主官网为准