跳转至

Phase 2: Roles and responsibilities（第二阶段：角色与职责）

The following are new roles and responsibilities for Phase 2 of the Foundry Program team, in addition to the roles established in Phase 1.

Education & Training Expert

Responsibilities

  • Coordinate training events that are aligned with business and use case prioritization; working with Use Case Leads and Domain Leads to align training with usse case launches, feature launches, and the addition of new cohorts of users to a given workflow.
  • Work with use case teams to develop training documentation specific to use cases and workflows, with the goal of ensuring that users and developers alike can understand different use cases and workflows and that training and documentation around those use cases meets program standards.
  • Interface with company-specific internal HR processes to include Foundry in a company's education program, certification, and execution; the Education & Training Expert should work with HR to integrate Foundry training and education into onboarding and continuing education programs across all organization verticals. Doing so will prevent the Foundry Program from becoming siloed and not integrated with the rest of the organization.
  • Evangelize Foundry through education; own the communication and publicization of available trainings and existing use cases. This can take many shapes depending on the organization structure but can include working with your organization's Learning and Development team to include Foundry training in development and upskilling opportunities, or working with internal communications or events teams to announce new training and certification opportunities and communicating the benefits of Foundry education.
  • Conduct skills gap analyses and audits for continuous improvement of training and education resources; develop assessments of the success of different training and education materials based on business outcomes and ROI improvements.
  • Create Foundry training programs to mark milestones and success criteria for different roles and use cases; these can be workshops, online trainings, self-service trainings, Palantir-provided learning resources, live trainings, and quizzes and documentation, as well as designing training projects in Foundry.
  • If a "Train-the-Trainer" program exists within one's organization, this role should coordinate and facilitate the delivery of this program; this allows Foundry education to expand beyond the constraints of just the dedicated education and training resource.

Profile and required skills

  • Knowledge of instructional design best practices and training methods, including coaching, workshops, and e-learning methods
  • Excellent communication and collaboration skills
  • Strong organizational skills and previous experience in project management
  • Some technical domain knowledge
  • Understanding of the business's structures and teams, and ability to liaise across teams and deeply understand their goals

Relevant Foundry applications and resources

Data Governance Administrator

Responsibilities

  • Focus on making data usable by users, in compliance with internal and external data quality standards and SLAs.
  • Own the documentation and maintenance of data governance requirements at the Program, and communicate these across workstreams.
  • Build monitoring and metrics around data governance processes, and own the continuous reporting and adjustment of these systems.
  • Partner with Permissions Managers and the Data Lead to implement groups and roles in project permissions. Create documentation and repeatable workflows across common project types and workflows to enable self-service on data governance requirements as applied to each project type.
  • Implement data governance review processes across all workstreams within the Foundry Program. Work with team leads to understand their workstream's needs and processes, and implement data governance checks, requirements, and processes to ensure data governance standards are being met across the Program.
  • If needed, interface with Palantir teams for support services delivery; leverage Palantir's resources, including documentation, advisory teams, and support services, to implement a data governance program in alignment with the organization's needs and with adequate vendor support systems in place.

:::callout{theme="neutral"} Within Phase 1, the Data Governance Administrator role can often be assumed by the Data Lead or the Ontology Lead. However, we recommend assigning this as a standalone role with dedicated resourcing by Phase 3 at the latest due to the scale of data management and complexity of workflows and use cases. :::

Profile and required skills

  • Familiarity with organization-wide data access permissions, regulations, and authentication systems (Okta and Active Directory, for example)
  • Programming skills (SQL, Python, PySpark/Spark)
  • Knowledge of data quality and version control best practices
  • Familiarity with agents / VMs
  • Strong technical communicator
  • Knowledge of systems architecture

Relevant Foundry applications and resources

Data Source Developer(s)

Responsibilities

  • Data source Developers are engineers who take on individual contributor tasks from the Data Lead and own the sourcing and cleaning process for all relevant data sources used across the Foundry Program.
  • We recommend assigning this role to a centralized team of Data Source Developers rather than decentralizing ownership among use case teams to maintain alignment with compliance, governance, and security requirements across all pipelines. Doing so also ensures standardization of data structures and pipeline architecture.
  • Each Data Source Developer owns the data connection configuration and technical connection to a set of data sources along with the Foundry data source Project, where the ingested data is made available in raw format.
  • As a team, Data Source Developers should own the creation of shared libraries and other documentation for data cleaning, cleaning automation, and data obfuscation when necessary.
  • Work with data owners in the organization to gather technical details about sources and to design pipelines that will implement data ingestion to best support organization goals and the end-state workflow's strategy and requirements.
  • Own daily management of data ingestion, pipeline health, and data quality. Ensure pipeline design, building, updating, and monitoring work is appropriately scheduled in alignment with Program goals and with use case and workflow requirements.
  • Manage and address failures or issues in data sources. This includes working with relevant IT teams and the Program Manager to create escalation SOPs, and monitoring and alerting systems for reactive response to failures, driving communications around failures, implementing remediation when failures arise, and creating retrospective processes to drive lessons learned from failures.
  • Focus on making data available to users, in compliance with internal and external data quality standards and SLAs. Work closely with Compliance teams and the Head of Data Governance to create guidelines, best practices, and review and monitoring processes for data pipelines.

Profile and required skills

  • Data Engineering skills (SQL, Python, PySpark/Spark)
  • Knowledge of data cleaning and quality best practices
  • Knowledge of version control best practices
  • Familiarity with agents / VMs / connecting systems

Relevant Foundry applications and resources

Ontology Manager(s)

Responsibilities

  • Own the development of the ontology from the existing model and production pipelines that then services as the consumption point for all workflow projects.
  • Manage the Ontology backlog, prioritizing change requests and triaging to other teams when required. Evaluate all requests for new objects and other changes to the Ontology with a clear view of the impact to the broader organizational Ontology and the evolution and roadmap of the Ontology in the long-term.
  • Partner with Use Case Leads and Data Source Developers on requests for new data sources to be ingested into Foundry; take a central role in all new use case architecture and data sourcing discussions to ensure alignment with the existing Ontology and with the Program-wide strategy for Ontology growth and evolution.
  • Own Ontology maintenance and develop a deep understanding of relevant data sources for each use case. Collaborate with Data Source Developers to understand data sources and become subject matter experts to plan integration into the ontology, and work with use case teams to understand the impact of Ontology design and decisions.
  • Work with Data Governance and Compliance teams to ensure access to the ontology and data sources are implemented in compliance with internal governance procedures. Develop scalable processes for governance checks in Ontology development and maintenances.
  • Work closely with the Ontology Lead and Data Lead to ensure alignment between the Ontology and data ingestion and maintenance processes. Develop the Ontology in alignment with the Program-wide roadmap and long-term strategy.

:::callout{theme="neutral"} The Ontology Manager role is often centralized, but for a mature platform (Phases 3 and 4) this role can be segmented into individual domains. In Phase 4, this role will provide guidance and feature requests to Pipeline Developers and Object View Developers to ensure the Ontology and data asset can best support ad hoc usage and use case development. :::

Profile and required skills

  • Data pipeline development and management skills
  • Database knowledge
  • Organizational domain knowledge
  • Capability to communicate between technical and non-technical stakeholders

Relevant Foundry applications and resources

中文翻译

第二阶段：角色与职责

以下是在第一阶段已确立的角色基础上，为 Foundry 项目团队第二阶段新增的角色与职责。

教育与培训专家

职责

  • 协调与业务和用例优先级一致的培训活动；与用例负责人和领域负责人合作，使培训与用例发布、功能发布以及向特定工作流添加新用户群体保持同步。
  • 与用例团队合作，针对具体用例和工作流开发培训文档，目标是确保用户和开发人员都能理解不同的用例和工作流，并且围绕这些用例的培训和文档符合项目标准。
  • 与公司特定的内部人力资源流程对接，将 Foundry 纳入公司的教育计划、认证和执行中；教育与培训专家应与人力资源部门合作，将 Foundry 的培训和教育整合到所有组织垂直领域的入职培训和持续教育项目中。这样做可以防止 Foundry 项目变得孤立，无法与组织其他部分融合。
  • 通过教育推广 Foundry；负责沟通和宣传可用的培训及现有用例。根据组织结构的不同，这可以采取多种形式，但可能包括与组织的学习与发展团队合作，将 Foundry 培训纳入发展和技能提升机会中，或与内部沟通或活动团队合作，宣布新的培训和认证机会，并传达 Foundry 教育的好处。
  • 进行技能差距分析和审计，以持续改进培训和教育资源；根据业务成果和投资回报率（ROI）的提升，评估不同培训和教育材料的成功程度。
  • 创建 Foundry 培训项目，为不同角色和用例设定里程碑和成功标准；这些可以是研讨会、在线培训、自助培训、Palantir 提供的学习资源、现场培训、测验和文档，以及在 Foundry 中设计培训项目。
  • 如果组织内存在“培训师培训”项目，此角色应协调并促进该项目的交付；这可以使 Foundry 教育超越仅由专职教育和培训资源负责的限制。

简介与所需技能

  • 了解教学设计最佳实践和培训方法，包括辅导、研讨会和电子学习方法
  • 出色的沟通和协作能力
  • 强大的组织能力和项目管理经验
  • 一定的技术领域知识
  • 理解业务结构和团队，能够跨团队联络并深入理解其目标

相关的 Foundry 应用和资源

数据治理管理员

职责

  • 专注于使用户能够使用数据，同时遵守内部和外部的数据质量标准及服务水平协议（SLA）。
  • 负责在项目层面记录和维护数据治理要求，并在各个工作流之间进行沟通。
  • 围绕数据治理流程构建监控和指标，并负责这些系统的持续报告和调整。
  • 与权限管理器和数据负责人合作，在项目权限中实施组和角色。针对常见的项目类型和工作流，创建文档和可重复的工作流，以便在应用于每个项目类型时，能够自助满足数据治理要求。
  • 在 Foundry 项目的所有工作流中实施数据治理审查流程。与团队负责人合作，了解其工作流的需求和流程，并实施数据治理检查、要求和流程，以确保整个项目满足数据治理标准。
  • 如有需要，与 Palantir 团队对接以获取支持服务交付；利用 Palantir 的资源，包括文档、咨询团队和支持服务，根据组织的需求并配备适当的供应商支持系统，来实施数据治理项目。

:::callout{theme="neutral"} 在第一阶段，数据治理管理员的角色通常可以由数据负责人或本体负责人兼任。然而，考虑到数据管理的规模以及工作流和用例的复杂性，我们建议最迟在第三阶段将其分配为一个独立的角色并配备专职资源。 :::

简介与所需技能

  • 熟悉组织级数据访问权限、法规和认证系统（例如 Okta 和 Active Directory）
  • 编程技能（SQL、Python、PySpark/Spark）
  • 了解数据质量和版本控制最佳实践
  • 熟悉代理/虚拟机
  • 强大的技术沟通能力
  • 了解系统架构

相关的 Foundry 应用和资源

数据源开发人员

职责

  • 数据源开发人员是承担数据负责人分配的个人贡献者任务的工程师，负责 Foundry 项目中所有相关数据源的获取和清洗过程。
  • 我们建议将此角色分配给一个集中的数据源开发人员团队，而不是分散给各个用例团队，以便在所有管道中保持与合规、治理和安全要求的一致性。这样做还可以确保数据结构和管道架构的标准化。
  • 每个数据源开发人员负责一组数据源的数据连接配置和技术连接，以及 Foundry 数据源项目，在该项目中，摄取的数据以原始格式提供。
  • 作为一个团队，数据源开发人员应负责创建共享库和其他文档，用于数据清洗、清洗自动化以及在必要时进行数据混淆。
  • 与组织中的数据所有者合作，收集有关数据源的技术细节，并设计用于实现数据摄取的管道，以最好地支持组织目标和最终工作流的策略及要求。
  • 负责数据摄取、管道健康和数据质量的日常管理。确保管道设计、构建、更新和监控工作根据项目目标以及用例和工作流要求进行适当安排。
  • 管理和处理数据源中的故障或问题。这包括与相关 IT 团队和项目经理合作，创建升级标准操作程序（SOP），以及用于故障被动响应的监控和警报系统，推动故障沟通，在故障发生时实施补救措施，并创建回顾流程以从故障中吸取教训。
  • 专注于使用户能够使用数据，同时遵守内部和外部的数据质量标准及服务水平协议（SLA）。与合规团队和数据治理负责人密切合作，为数据管道创建指南、最佳实践以及审查和监控流程。

简介与所需技能

  • 数据工程技能（SQL、Python、PySpark/Spark）
  • 了解数据清洗和质量最佳实践
  • 了解版本控制最佳实践
  • 熟悉代理/虚拟机/连接系统

相关的 Foundry 应用和资源

本体管理器

职责

  • 负责从现有模型和生产管道开发本体，该本体随后作为所有工作流项目的消费点。
  • 管理本体待办事项列表，对变更请求进行优先级排序，并在需要时将其分流给其他团队。评估所有新对象和对本体的其他更改请求，并清晰了解其对更广泛组织本体的影响以及本体的长期演进路线图。
  • 与用例负责人和数据源开发人员合作，处理将新数据源摄取到 Foundry 的请求；在所有新用例架构和数据源讨论中发挥核心作用，以确保与现有本体以及项目范围内本体增长和演进策略保持一致。
  • 负责本体维护，并深入了解每个用例的相关数据源。与数据源开发人员合作，理解数据源并成为主题专家，以规划其集成到本体中，并与用例团队合作，理解本体设计和决策的影响。
  • 与数据治理和合规团队合作，确保对本体和数据源的访问符合内部治理程序。为本体开发和维护中的治理检查开发可扩展的流程。
  • 与本体负责人和数据负责人密切合作，确保本体与数据摄取和维护流程保持一致。根据项目范围内的路线图和长期策略开发本体。

:::callout{theme="neutral"} 本体管理器角色通常是集中式的，但对于一个成熟的平台（第三和第四阶段），此角色可以细分为各个领域。在第四阶段，此角色将为管道开发人员和对象视图开发人员提供指导和功能请求，以确保本体和数据资产能够最好地支持临时使用和用例开发。 :::

简介与所需技能

  • 数据管道开发和管理技能
  • 数据库知识
  • 组织领域知识
  • 能够在技术和非技术利益相关者之间进行沟通

相关的 Foundry 应用和资源