Senior Software Engineer, NeMo Core Platform

免费注册后解锁"你的匹配分 vs 高分画像"与个人技能缺口。

We are looking for a Senior Software Engineer to help build NeMo Helix , NVIDIA ’s product for developing, evaluating, deploying, and operating AI systems at scale. This role is for a senior engineer/architect for our Core team which owns and ships an open source plugin-based AI platform for running and optimizing Agents targeting multiple compute backends (local/docker, Kubernetes, Slurm, etc.). As AI systems become more autonomous and more deeply integrated into real workflows, teams need robust APIs and orchestration systems for running, monitoring, and optimizing Agents at scale. Increasingly, the users of these systems are themselves autonomous or semi-autonomous Agents. The NeMo Helix group is building a sophisticated agent execution framework to enable agents to automatically run hundreds of experiments in parallel to find the most efficient agent architectures for our customers. This is important product engineering research for making agents more sustainable. AI systems are not yet nearly as efficient as they can be, and systems like NeMo Helix will allow large scale AI consumers to automatically tune their agents to use fewer tokens and rely on more efficient models with better throughput. What you'll be doing: Working in a product research environment where we place big bets on where the future is heading, adapting in real time as we build alongside an industry that is constantly evolving with us. This means fast iteration, high ownership, pragmatic decisions, and performance-minded implementation under production constraints Designing an Agentic Execution system that are flexible enough to work in many environments (local, k8s, Slurm, on prem / air-gapped) Provide senior technical leadership through design reviews, code reviews, mentoring, and ownership of ambiguous cross-component problems Building and maintain our Core Platform APIs for running jobs, storing data, entities, secrets, and RBAC and Auth Extending our flexible Plugin Architecture that makes it easy for many teams and external customers to install new capabilities into our system Building in the open in our OSS repo, keeping up the high standards that the open source community demands Shipping code at the speed of light with an unlimited token budget using best in class agentic coding tools Improving reliability, observability, debuggability, and performance across NeMo Helix, SDKs, plugins, jobs, and developer workflows Building strong test coverage across unit, integration, E2E, Docker, and Kubernetes workflows What we need to see: BS, MS, or equivalent experience in Computer Science, Computer Engineering, or a related technical field 10+ years of professional software engineering experience building production systems Comfort working in a very fast and ambiguous environment Exceptional communication, both verbal and written. This includes the ability to produce and review high quality architectural RFCs, and to discuss them clearly with the right level of technical detail for the right people (Engineer, Product, Marketing, etc.) Strong system design skills, with a pragmatic flexibility and phenomenal instincts to invent robust systems quickly without over-complicating. Strong understanding of reliability, scalability, security, and performance tradeoffs in production infrastructure Experience with distributed systems, cloud-native services, containers, Kubernetes, and job orchestration Excellent Python engineering skills, including API design, typing, testing, debugging, performance analysis, and maintainable software design Experience designing SDKs, libraries, plugins, CLIs, or other developer-facing interfaces Ability to work independently, define technical scope, break down ambiguous problems, and drive work across team boundaries Ways to stand out from the crowd: Experience building, deploying, and iterating on production agentic AI systems at scale in Kubernetes Experience with sophisticated plugin architectures Strong ability to connect tech

我们正在寻找一位高级软件工程师，帮助构建 NeMo Helix——NVIDIA 用于大规模开发、评估、部署和运营 AI 系统的产品。该职位面向我们 Core 团队的高级工程师/架构师，该团队拥有并交付一个基于插件的开源 AI 平台，用于运行和优化面向多种计算后端（本地/docker、Kubernetes、Slurm 等）的 Agents。随着 AI 系统变得更加自主并更深入地集成到实际工作流中，团队需要强大的 API 和编排系统来大规模运行、监控和优化 Agents。越来越多的这些系统的用户本身就是自主或半自主的 Agents。NeMo Helix 团队正在构建一个复杂的 agent 执行框架，使 agents 能够自动并行运行数百个实验，为客户找到最高效的 agent 架构。这是一项重要的产品工程研究，旨在让 agents 更具可持续性。AI 系统尚未达到其应有的效率，而像 NeMo Helix 这样的系统将允许大规模 AI 消费者自动调优其 agents，以使用更少的 tokens，并依赖具有更好吞吐量的更高效模型。你将要做的事情：在产品研究环境中工作，我们对未来的发展方向做出重大押注，并与不断与我们共同演进的行业一起实时调整。这意味着快速迭代、高度主人翁意识、务实的决策，以及在生产约束下注重性能的实现。设计一个 Agentic Execution 系统，足够灵活以适用于多种环境（本地、k8s、Slurm、本地部署/气隙环境）。通过设计评审、代码评审、指导以及对模糊的跨组件问题的所有权，提供高级技术领导力。构建并维护我们的 Core Platform APIs，用于运行作业、存储数据、实体、密钥以及 RBAC 和 Auth。扩展我们灵活的 Plugin Architecture，使许多团队和外部客户能够轻松地将新能力安装到我们的系统中。在我们的 OSS 仓库中公开构建，保持开源社区所要求的高标准。使用业界一流的 agentic 编码工具，以光速交付代码，并拥有无限的 token 预算。提升 NeMo Helix、SDKs、plugins、jobs 和开发者工作流的可靠性、可观测性、可调试性和性能。在单元、集成、E2E、Docker 和 Kubernetes 工作流中构建强大的测试覆盖率。我们需要看到：计算机科学、计算机工程或相关技术领域的学士、硕士或同等经验。10 年以上构建生产系统的专业软件工程经验。能够在非常快速且模糊的环境中舒适工作。卓越的口头和书面沟通能力。这包括能够编写和评审高质量的架构 RFC，并以适合合适对象（工程师、产品、市场等）的恰当技术细节水平清晰地讨论它们。强大的系统设计能力，具备务实的灵活性以及非凡的直觉，能够快速发明稳健的系统而不过度复杂化。深入理解生产基础设施中的可靠性、可扩展性、安全性和性能权衡。具有分布式系统、云原生服务、容器、Kubernetes 和作业编排经验。卓越的 Python 工程能力，包括 API 设计、类型标注、测试、调试、性能分析和可维护的软件设计。具有设计 SDKs、库、插件、CLIs 或其他面向开发者的接口的经验。能够独立工作，定义技术范围，分解模糊问题，并推动跨团队边界的工作。脱颖而出的方式：具有在 Kubernetes 中大规模构建、部署和迭代生产级 agentic AI 系统的经验。具有复杂插件架构的经验。强大的能力将技术

⚑ 岗位信息收录自公开招聘页（Himalayas），以雇主官网为准