A Juejin tutorial maps the first 19 chapters of its RAG series onto an enterprise knowledge-base question-answering architecture. It combines hybrid dense and sparse retrieval, reranking, Self-Reflective RAG, agent queries, streaming responses, evaluation, observability, and feedback loops, with exercises for implementation and validation.
A technical article uses a teaching scenario in which an agent cites the correct announcement but reports an incorrect number, with no API error and an all-green tracing page. It proposes four checkpoints: source to tool result, tool result to actual model request, request to generated answer, and generated answer to user-visible text. The article also draws on the authors’ own Langfuse observability records, covering a 256 KiB body-capture limit, a double-slash URL that redirected requests and caused a background export error, an isolated ClickHouse recovery drill, and de-duplication of cost records representing the same physical request. The opening scenario is explicitly a teaching exercise; possible causes such as missing units, omitted footnotes, incomplete input and model misunderstanding are branches rather than a confirmed root cause.
What problem it solves
Supported Echo Show devices otherwise run Fire OS and Alexa, tying their voice and display functions to Amazon services. TECHO5 replaces those components with Alpine Linux and a Go daemon so the device can operate as a local Home Assistant voice satellite.
Key features
Home Assistant voice satellite — The firmware connects supported displays to Home Assistant as voice satell…
What problem it solves
Teams often spread project context and decisions across chat, source-control tools, CI dashboards, release tools, search systems, and automation. Buzz places these activities in one event log so messages, workflows, code changes, reviews, and approvals share an identity model and audit trail.
Key features
Shared rooms for humans and agents — People and AI agents participate …
What problem it solves
Teams need to bring documents together so they can be searched, reasoned over and kept up to date. They also need a way to move from source-grounded retrieval to multi-step tasks while maintaining organized, current knowledge.
Key features
Three knowledge workflows — RAG supports lookup, an agent handles multi-step tasks, and a wiki organizes knowledge. All three operate on …