Equinox is an F# library for event-sourced systems, providing a consistent abstraction over multiple stream-based storage backends including CosmosDB, DynamoDB, EventStoreDB, MessageDB, SqlStreamStore, and an in-memory store. In production, it enables applications to define domain logic as pure decision functions that produce events, while the library manages optimistic concurrency, caching, snapshots, and retries. The core is the `Decider` pattern: an aggregate loaded from a stream, a decide function that returns events, and a fold that applies events to produce new state. Access strategies (e.g., `LatestKnownEvent`, `RollingSnapshots`, `Custom`) control how state is reconstructed from stored events, trading read cost for write cost. The architecture is modular: the `Equinox` package defines the core interfaces (`ICategory`, `IStream`, `Decider`), while separate NuGet packages implement storage adapters. The library is used as a dependency in larger applications; sample projects (`Store/Domain`, `TodoBackend`, `Tutorial`) demonstrate its use. Key entry points include the `Decider` class for command/query execution, `Category` for resolving streams, and store-specific connectors (e.g., `CosmosStoreConnector`, `DynamoStoreConnector`). A CLI tool (`Equinox.Tool`) provides provisioning, benchmarking, and data inspection commands across all stores. External dependencies are minimal per package: `Equinox` depends only on `FsCodec` for serialization; store adapters require their respective SDKs (Azure.Cosmos, AWSSDK.DynamoDBv2, EventStore.Client, Npgsql, etc.). The library includes comprehensive caching (`Equinox.Cache` wrapping `System.Runtime.Caching.MemoryCache`), metrics via Prometheus, and a load testing harness. Despite some deprecated adapters (legacy EventStore TCP, SqlStreamStore) and high complexity from multiple backends, the core patterns are clean, well-tested, and production-proven at Jet.com.
This is a sample application from Microsoft demonstrating how to build a chatbot with Azure OpenAI and your own data using Retrieval-Augmented Generation (RAG). It is a full-stack web application with a Python Quart async backend and a React/TypeScript frontend. The backend handles conversation orchestration, integrates with Azure OpenAI for chat completions, and supports multiple data sources (Azure Cognitive Search, CosmosDB, Elasticsearch, Pinecone, etc.) for grounding responses. It also includes a comprehensive data preparation pipeline (scripts) for chunking documents, generating embeddings, and indexing into search services. The frontend provides a modern chat interface with features like streaming responses, citation display, chat history, and file upload. The repository is primarily an application but has a strong data engineering component, making it a hybrid. It is actively maintained, well-structured with tests and CI/CD, but has some technical drift due to preview APIs and outdated dependencies.