Search for a command to run...
Articles tagged with #systemdesign
The ML system design framework: framing the task, offline vs online metrics, feature stores and train-serve skew, candidate generation plus ranking, and drift monitoring.
How a time-series database handles massive append-heavy writes: the series model, columnar storage, Gorilla compression, high cardinality, downsampling, and retention.
How a distributed key-value store like DynamoDB stays fast at any scale: consistent hashing, quorum replication, hot partitions, conflict resolution, global tables.
Retrieval-augmented generation is a retrieval problem wearing a language model's clothes. Chunking, embeddings, ANN, hybrid retrieval, reranking, and eval.
How real-time chat systems like Slack push messages to millions of connected clients instantly. Covers persistent WebSocket connections, separating the edge from message logic, pub/sub fanout, and delivery guarantees.
Two ideas explain why full-text search is fast at scale: the inverted index and shard routing. Covers how documents are indexed, how queries scatter and gather, and the consistency trade-offs.
