⏱ 9 min read A vector database stores high-dimensional embedding vectors and enables fast approximate nearest neighbor (ANN) search — finding the top-K vectors […] Read article

⏱ 9 min read A time-series database (TSDB) is purpose-built for append-heavy, time-ordered metric data. Its key design decisions — chunk storage, timestamp and Read article

⏱ 8 min read A graph database stores entities as nodes and relationships as edges, with properties on both. Unlike relational databases that simulate Read article

⏱ 8 min read A column-family store organises data in a hierarchy: keyspace → table → partition → rows. The partition key determines which Read article

⏱ 12 min read Key-Value Store System: Overview A key-value store is the simplest distributed database: keys map to opaque binary values. Despite its Read article

⏱ 11 min read Document Store System: Overview A document store persists and retrieves arbitrary JSON documents. Unlike relational databases, it does not enforce Read article

⏱ 8 min read Synonym Expansion System: Overview Synonym expansion augments search queries with equivalent or related terms so that a search for “automobile” Read article

⏱ 9 min read Spell Checker System: Overview A spell checker takes an input word (potentially misspelled) and returns a ranked list of correction Read article

⏱ 9 min read Typeahead Autocomplete System: Overview A typeahead (autocomplete) system returns ranked query completions as the user types, with latency under 50ms Read article

⏱ 8 min read Proximity Search System: Overview A proximity search system finds entities (restaurants, drivers, users) within a geographic radius of a query Read article

⏱ 8 min read Graph Search System: Overview A graph search system enables queries over a network of nodes and edges — users, connections, Read article

⏱ 7 min read Recommendation Feedback Loop: Overview A recommendation system without a feedback loop becomes stale. The feedback loop closes the cycle: collect Read article

⏱ 8 min read User Activity Feed: Overview A user activity feed aggregates events from actors a user follows and surfaces them in reverse-chronological Read article

⏱ 9 min read Why a Dedicated Experiment Platform? A/B testing sounds simple — split users, measure a metric, run a t-test — but Read article

⏱ 8 min read Why Model Serving Is Not Trivial Deploying a model to production is an engineering discipline separate from training one. A Read article