MZ Marshall - 拥抱变化 Marshall - Embrace Change

我是张洋（Marshall），专注 AI 与平台产品管理的产品经理。 I am Marshall Zhang, a product manager focused on AI and platform product management.

我具备分布式数据系统执行平台建设经验，当前负责 Microsoft Fabric 执行平台核心能力设计与落地。我关注如何把调度、监控、执行身份和 AI 能力整合进真实工作流，形成可规模化、可依赖、可衡量的产品价值。 I build execution platforms for distributed data systems, with current ownership across core Microsoft Fabric job execution experiences. I focus on turning scheduling, observability, execution identity, and AI into scalable, trustworthy, measurable product value.

查看职业路径 View career path 联系我 Contact me GetThatJob! Try GetThatJob!

职业经历 Career experience

从分布式系统工程到 AI 驱动的平台产品。 From distributed systems engineering to AI-driven platform products.

职业路径 Career path

具备调度、监控、执行平台建设经验，能在性能、可靠性与易用性之间做产品级取舍。 A product manager experienced in scheduling, observability, and execution platforms, with strong judgment across performance, reliability, and usability tradeoffs.

我的经历从 Hadoop、数据安全和基础设施工程开始，逐步进入 Azure Synapse 与 Microsoft Fabric 的平台产品设计。当前我重点负责 Fabric Job Platform 相关能力，并持续探索 AI 在工作流自动化、智能监控和产品研发效率中的真实落地价值。 My path started in Hadoop, data security, and infrastructure engineering, then moved into platform product work across Azure Synapse and Microsoft Fabric. Today I own Fabric Job Platform capabilities and explore how AI can create real value in workflow automation, intelligent monitoring, and product development productivity.

工作经历 Experience

2022.03 - 至今 Mar 2022 - Present

微软 · 北京 Microsoft · Beijing

产品经理 · C+AI / Microsoft Fabric Program Manager · C+AI / Microsoft Fabric

主导 Monitoring Hub 统一观测平台，解决 Fabric 不同工作负载之间监控入口割裂、排查链路分散的问题。通过聚焦任务发现、故障诊断和跨工作负载调试，推动产品达到 40万+ MAU，约占 Fabric MAU 的 80%。 Drove Monitoring Hub as a unified observability platform for Fabric, addressing fragmented monitoring and troubleshooting across workloads. Scaled adoption to 400K+ MAU, about 80% of Fabric MAU, by focusing on job discovery, failure diagnosis, and cross-workload debugging.

面向企业级数据流水线，重构调度与并发模型，支持单用户每分钟 1000+ 执行和长周期任务；在 ADF / HDInsight 向 Fabric 迁移过程中，解决并发限制、API 限流和可靠性等关键执行瓶颈。 Redefined scheduling and concurrency models for enterprise-scale pipelines, supporting 1000+ executions per minute per user and long-running jobs. Unblocked ADF / HDInsight to Fabric migration by resolving concurrency, API throttling, and reliability constraints.

推动 AI 驱动的工作流与效率提升，包括用户反馈智能分流、失败摘要、模式聚类、异常检测，以及覆盖 PM / 设计 / 开发的 AI 辅助研发流程，将部分文档与设计迭代周期从周级缩短到天级。 Drove AI-assisted workflow improvements, including customer feedback triage, failure summarization, pattern clustering, anomaly detection, and an AI-assisted PM/design/engineering delivery pipeline that reduced selected documentation and design cycles from weeks to days.

Monitoring Hub Job Scheduler AI Workflow Execution Scalability
2021.03 - 2022.05 Mar 2021 - May 2022

微软 · 北京 Microsoft · Beijing

软件工程师（前端）· Microsoft Fabric Software Engineer (Frontend) · Microsoft Fabric

参与 Fabric 平台核心前端体验建设，负责 Homepage、Playground 等关键界面，优化用户交互体验，并构建可复用组件。 Developed core frontend experiences for the Fabric platform, including Homepage and Playground, improving usability and interaction quality while building reusable components.

参与端到端测试、可访问性合规和单元测试实践，推动产品体验与工程质量同步提升。 Contributed to end-to-end testing, accessibility compliance, and unit testing practices to improve both product experience and engineering quality.

React TypeScript Accessibility E2E
2020.03 - 2021.05 Mar 2020 - May 2021

微软 · 北京 Microsoft · Beijing

软件工程师（前端）· Azure Streaming Software Engineer (Frontend) · Azure Streaming

完成 Azure Streaming 服务在 Azure Synapse 中的前端集成，并设计 DAG 可视化界面，帮助用户理解流式任务的执行关系。 Delivered UI integration for Azure Streaming Service within Azure Synapse and built a DAG-like visualization experience to represent streaming job execution relationships.

Angular TypeScript C# DAG UX
2019.01 - 2020.03 Jan 2019 - Mar 2020

微软 · Redmond Microsoft · Redmond

软件工程师（后端）· HDInsight Software Engineer (Backend) · HDInsight

参与 Hadoop 平台日志与任务调度系统开发，负责生产环境支持和值班（DRI），建立了对企业级数据平台基础设施的实践理解。 Worked on Hadoop infrastructure, including logging systems and job queue management, while supporting production systems as an on-call engineer (DRI).

Hadoop Hive YARN Distributed Systems
2015.07 - 2018.12 Jul 2015 - Dec 2018

Dataguise · Fremont Dataguise · Fremont

软件工程师 / Scrum Master Software Engineer / Scrum Master

开发基于 Hadoop 的数据安全与合规平台，覆盖 GDPR 与敏感数据保护场景，同时负责项目管理、迭代推进与生产支持。 Developed Hadoop-based data security and compliance solutions for GDPR and sensitive data protection, while leading Scrum processes, project tracking, and production support.

Java Hadoop Data Security Agile

教育背景 Education

2013 - 2015

SUNY Binghamton University

电子与计算机工程硕士（M.S.）｜GPA 4.0 Master of EECE | GPA 4.0

2009 - 2013

北京邮电大学 Beijing University of Posts and Telecommunications

电子工程学士（B.S.）｜GPA 3.2 BS of EE | GPA 3.2

核心能力 Core capabilities

我希望把 AI 落到真实工作流和平台级产品里。 I want to bring AI into real workflows and platform-scale products.

平台产品判断 Platform product judgment

理解分布式执行、调度、监控、限流、可靠性和身份等平台能力，能在用户体验与系统稳定性之间做清晰取舍。 I understand distributed execution, scheduling, observability, throttling, reliability, and identity, and can make clear product tradeoffs between user experience and platform protection.

AI 落地能力 AI implementation depth

关注 AI 在真实系统中的价值边界：哪些任务适合自动化，哪些需要辅助决策，如何把 AI 融入端到端流程而不是堆成孤立功能。 I focus on the practical boundary of AI in real systems: what should be automated, what should assist human decisions, and how to integrate AI into end-to-end workflows rather than isolated features.

结果导向执行 Outcome-oriented execution

能从复杂信息中识别高价值场景、定义优先级并推动跨团队落地，最终用采用率、效率提升、可靠性和客户成功来衡量产品结果。 I identify high-impact scenarios from complex information, define priorities, and drive cross-team execution, measuring outcomes through adoption, efficiency gains, reliability, and customer success.

Contact Contact

如果你正在寻找 AI 或平台产品方向的人选，可以从这里开始。 If you are looking for AI or platform product talent, start here.

查看中文简历 View English resume