Data Lake Architect

# Data Lake Architect ## Overview **Data Lake Architect** designs large-scale data storage architectures for multi-structured data processing and analytics. It structures raw/curated/analytics zones, cataloging and lineage, governance and access control, schema evolution, and analytics performance optimization on cloud-native object storage. ## Core Capabilities - ✅ **Zone architecture** — raw/bronze → curated/silver → analytics/gold, with ownership and lifecycle rules - ✅ **Ingestion pipelines** — batch + streaming, idempotent, partitioned, columnar formats - ✅ **Catalog & metadata** — schema registry, lineage, ownership, classification, statistics - ✅ **Governance & security** — access control, PII masking, retention, audit - ✅ **Performance** — partitioning, clustering, file sizing, compaction, query-engine tuning - ✅ **Lifecycle & cost** — hot/warm/cold tiering, archival, deletion, cost model ## Usage 1. Load the `data-lake-architect` agent into the Markus workspace. 2. Provide the source systems, target analytics workloads, and existing storage/config artifacts. 3. The agent produces the architecture document, ingestion design, catalog/metadata model, governance policy, performance plan, and lifecycle/cost plan — registered via `deliverable_create`, with build stages tracked via `task_create` (using `blocked_by`). ## License & Source Based on [data-lake-architect](https://github.com/gensecaihq/Claude-Code-Subagents-Collection/blob/main/subagents/data-analytics/data-lake-architect.md) (MIT). --- **Attribution:** Based on [data-lake-architect](https://github.com/gensecaihq/Claude-Code-Subagents-Collection/blob/main/subagents/data-analytics/data-lake-architect.md) from [Claude-Code-Subagents-Collection](https://github.com/gensecaihq/Claude-Code-Subagents-Collection) - Original author: gensecaihq contributors - License: MIT