Lead AI Engineer - NBA

免费注册后解锁"你的匹配分 vs 高分画像"与个人技能缺口。

Become a part of our caring community The Lead AI Engineer owns the design and delivery of the NBA platform's agentic AI capabilities. This is a hands-on engineering role first - you are expected to be a significant individual contributor building production AI systems every sprint. You also lead a small team of engineers and contractors focused on bringing AI-powered workflows and intelligent analysis into the platform's user-facing experiences. This role is forward-looking: you are building the foundation for how AI augments and automates decision-making across the platform. What You'll Own Agentic AI delivery - design, build, and own AI agent workflows that surface intelligent analysis, recommendations, and automation across the NBA platform's user interfaces and internal tooling Hands-on development - actively write and ship production code alongside your team; this is not a supervisory role AI workflow design - architect multi-step agentic pipelines that integrate with the platform's data layer, decisioning outputs, and activation channels to deliver meaningful AI-driven experiences LLM integration - own the integration of large language models (Claude, GPT, or equivalents) into platform workflows, including prompt engineering, tool use, structured output handling, and context management User-facing AI experiences - partner with frontend and product teams to bring AI-powered analysis and recommendations into the Command and Control UI and other operator-facing surfaces Evaluation and reliability - build evaluation frameworks to measure AI output quality; ensure agentic workflows are reliable, auditable, and safe in a regulated healthcare environment Contractor management - manage AI engineering contractors within your pod; set expectations, review work, and hold the team to quality standards AI productivity tooling - model best practices for AI-assisted development (Claude, Copilot, etc.) across the broader engineering organization Use your skills to make an impact Required Qualifications 6+ years of software or ML engineering experience with at least 1-2 years focused on LLM integration, agentic AI, or AI product engineering Hands-on experience building production agentic systems - multi-step pipelines, tool orchestration, structured output handling, and multi-turn conversation management Experience with RAG patterns and vector search - you have built retrieval-augmented systems in production, not just read about them Strong Python engineering skills; comfortable building and shipping production-grade AI systems, not just prototypes Experience integrating LLM APIs (Anthropic, OpenAI, or equivalents) into real applications Proven ability to lead a small team or pod through delivery and manage contractor resources Strong judgment on when AI is the right tool and when it is not - especially in a regulated, member-facing context Clear communicator - you can explain an agentic architecture to an engineer and a product owner without losing either of them Deep fluency with AI productivity tools (Claude, Copilot, or similar) - this is table stakes for this role Preferred Qualifications Experience orchestrating multi-agent systems - coordinating multiple AI agents across complex, long-running workflows Experience integrating agentic AI into bespoke enterprise software and connecting to external systems like Salesforce, Adobe Experience Platform, claims platforms, or similar Experience building AI-powered analytics or insight surfaces in a business intelligence or operational UI context Background in healthcare, insurance, or other regulated industries where AI outputs carry compliance risk Experience with AI evaluation frameworks and techniques for measuring output quality and detecting drift Familiarity with the Anthropic Claude API and the Claude Agent SDK. Work Hours : Typical business hours are Monday-Friday, 8 hours/day, 5 days/week-- some flexibility might be possible, depending on business needs. Very minimal travel might

加入我们充满关怀的社区。首席AI工程师负责设计并交付NBA平台的代理式AI能力。这首先是一个动手工程岗位——你需要作为重要的个人贡献者，每个冲刺都构建生产级AI系统。你还要带领一支由工程师和承包商组成的小团队，专注于将AI驱动的工作流和智能分析引入平台的面向用户体验。这个岗位具有前瞻性：你正在为AI如何在整个平台上增强和自动化决策奠定基础。你将负责什么 代理式AI交付——设计、构建并负责AI代理工作流，在NBA平台的用户界面和内部工具中呈现智能分析、建议和自动化 动手开发——与团队一起积极编写并交付生产代码；这不是一个监督性岗位 AI工作流设计——架构多步骤代理式流水线，与平台的数据层、决策输出和激活渠道集成，以交付有意义的AI驱动体验 LLM集成——负责将大型语言模型（Claude、GPT或同等产品）集成到平台工作流中，包括提示工程、工具使用、结构化输出处理和上下文管理 面向用户的AI体验——与前端和产品团队合作，将AI驱动的分析和建议引入Command and Control UI及其他面向操作员的界面 评估与可靠性——构建评估框架以衡量AI输出质量；确保代理式工作流在受监管的医疗保健环境中可靠、可审计且安全 承包商管理——管理你所在小组中的AI工程承包商；设定期望、审查工作，并让团队达到质量标准 AI生产力工具——在更广泛的工程组织中示范AI辅助开发（Claude、Copilot等）的最佳实践 运用你的技能创造影响力 必备资格 6年以上软件或ML工程经验，其中至少1-2年专注于LLM集成、代理式AI或AI产品工程 具备构建生产级代理式系统的动手经验——多步骤流水线、工具编排、结构化输出处理和多轮对话管理 具备RAG模式和向量搜索经验——你在生产环境中构建过检索增强系统，而不只是读过相关材料 扎实的Python工程技能；能够构建并交付生产级AI系统，而不只是原型 具备将LLM API（Anthropic、OpenAI或同等产品）集成到真实应用中的经验 具备带领小团队或小组完成交付并管理承包商资源的 proven 能力 对AI何时是正确工具、何时不是具有强判断力——尤其是在受监管、面向会员的语境中 清晰的沟通者——你能向工程师和产品负责人解释代理式架构，同时不让任何一方掉队 深度熟练使用AI生产力工具（Claude、Copilot或类似工具）——这是该岗位的基本要求 优先资格 具备编排多代理系统的经验——在复杂、长时间运行的工作流中协调多个AI代理 具备将代理式AI集成到定制企业软件并连接外部系统（如Salesforce、Adobe Experience Platform、理赔平台或类似系统）的经验 具备在商业智能或运营UI语境中构建AI驱动分析或洞察界面的经验 具备医疗保健、保险或其他受监管行业的背景，在这些行业中AI输出带有合规风险 具备AI评估框架以及衡量输出质量和检测漂移技术的经验 熟悉Anthropic Claude API和Claude Agent SDK。工作时间：典型工作时间为周一至周五，每天8小时，每周5天——根据业务需求可能有一定灵活性。极少量出差可能

⚑ 岗位信息收录自公开招聘页（Himalayas），以雇主官网为准