A ubiquitous public account for pirate enthusiasts, regularly sharing curated experiences, tech learning, and growth insights. Currently publishing articles on AI RAG customer service, AI MCP technology, and open-source design. Personal free Knowledge Planet: Awakening New World Programmer.
This article walks through building a FastGPT‑powered AI agent that automatically parses, scores, and records resumes in Feishu multi‑dimensional tables, detailing the problem of manual screening, the workflow configuration, prompt design, and the resulting efficiency gains.
This article explains why enterprises need a single gateway for multiple large‑model services, introduces the open‑source OneAPI tool and its features, and provides step‑by‑step Docker installation, UI setup, channel and token configuration, optional Nginx proxy, and integration into an AI code‑review project.
ByteDance’s internally built “塔阁平台” (Tag Platform) powers Douyin e‑commerce’s trillion‑scale tag management and audience selection, handling over a thousand tag types, millions of audience combinations, and ten‑million QPS through a hybrid offline‑online architecture that combines Hive, ClickHouse, Bitmap compression, rule‑engine and materialized views, while ensuring high availability, low latency, and robust monitoring.
This tutorial walks through creating a Sequential Thinking MCP Server in Java using Spring AI 1.0, covering configuration, core classes, tool registration, HTTP SSE setup, performance trade‑offs, and links to the official guide and GitHub repository.
This article explains how to enhance a travel‑focused AI customer‑service system by importing structured and unstructured data into a RAG pipeline, then applying anaphora resolution and query rewriting techniques—illustrated with FastGPT workflows, prompt templates, and practical examples—to improve retrieval accuracy and answer relevance.
The DeepSeek‑R1‑0528 update brings major benchmark gains, reduced hallucinations, JSON output and function‑calling support, enabling seamless integration with the Model Context Protocol (MCP) for flexible tool use, while a new 8B distilled model demonstrates knowledge‑distillation benefits for resource‑constrained deployments.
This article walks through using the open‑source workflow tool N8N to automatically fetch scenic‑spot data from the Ly.com travel site, transform it with an AI node, and ingest the structured knowledge into FastGPT, enabling an AI‑powered customer service chatbot without writing custom code.
This article explains why Retrieval‑Augmented Generation (RAG) is essential for AI‑driven customer service, outlines the limitations of large language models, details the three‑stage RAG workflow (indexing, retrieval, generation), and shows how to implement it with FastGPT, vector databases, and LangChain.
This article walks through converting the open‑source SequentialThinking MCP Server from its default Stdio interface to an HTTP Server‑Sent Events (SSE) implementation using TypeScript, Express, and the MCP SDK, detailing configuration, code changes, route setup, and testing procedures.
This article explains how Supergateway converts a standard Stdio‑based Model Context Protocol (MCP) server into an HTTP SSE (or WebSocket) service, detailing installation, command‑line usage, a train‑ticket MCP example, and how the conversion simplifies web integration and remote debugging.
