ddia-systems

The ddia-systems skill provides a structured framework for designing and reasoning about data-intensive applications. It guides users through key concepts in distributed systems, database selection, replication, partitioning, consistency, and transactions. By emphasizing the principle that data outlives code, this skill helps ensure that data architectures are durable, maintainable, and resilient over time. It addresses common challenges such as managing high data volumes, handling evolving schemas, and making explicit trade-offs between consistency, availability, and performance.

This skill offers capabilities in evaluating data models and query languages, understanding storage engine trade-offs (like LSM trees vs. B-trees), and assessing replication and partitioning strategies. It provides insights into batch and stream processing, distributed consensus mechanisms, and the effects of different consistency models on application behavior. Users can score data architectures on a scale from 0 to 10, identifying areas of accidental complexity, failure modes, and opportunities for improvement. The skill also highlights polyglot persistence, schema-on-write vs. schema-on-read approaches, and the importance of matching data models to specific use cases.

Target users include software engineers, system architects, and data engineers who design databases, build distributed systems, or implement complex data pipelines. It is especially useful for those making decisions between SQL and NoSQL solutions, optimizing for write-heavy or read-heavy workloads, and debugging issues related to replication lag or eventual consistency. The skill is practical for anyone involved in long-term system design where correctness, durability, and evolvability of the data layer are critical.