A platform for sharing and discussing tech knowledge, guiding you toward the cloud of technology.
The Dewu tech team’s recent “boundary‑crossing” incidents—engineers skipping formal specs to talk directly with users and operations—led to deeper problem understanding, rapid prototyping, and measurable improvements such as 60% automated answers, 30% support load reduction, and 50% faster onboarding.
This article explains why Retrieval‑Augmented Generation is needed, walks through the offline and online pipeline, details knowledge sources, embedding models, chunking strategies, similarity metrics, the HNSW index, multi‑path recall, RRF fusion and cross‑encoder reranking, and summarizes practical takeaways for building effective RAG systems.
The article details DeWu's evolution from manual, experience‑driven recommendation troubleshooting to an AI‑powered diagnostic platform called “PushCheck”, describing its dual‑mode Highway/ATV architecture, the Story‑Skill framework, knowledge‑base integration, evolution pipeline, and a real‑world case study.
DeWu, a leading fashion e‑commerce platform, evaluated and adopted OceanBase as a multi‑model database to replace a heterogeneous stack of MySQL, TiDB, MongoDB, HBase, ClickHouse, and others, achieving up to 200× query speedups, 40%+ storage savings, high‑availability with RPO = 0, and a streamlined operations model while documenting migration challenges and future roadmap.
The article analyzes the limitations of traditional UI automation, introduces the AI‑Native ai_uitester pipeline that converts test‑case data with LLM enhancement, implements AI‑driven debugging and self‑healing, and unifies cross‑platform execution through a VLM‑based engine, backed by real‑world metrics.
The article chronicles how a marketing activity that required ten system switches and over forty manual fields was transformed by replacing simple AI‑assisted form filling with a two‑stage Agent architecture and an aggregated workbench, detailing the architectural choices, trade‑offs, and practical lessons learned.
The article details how Hermes Agent transforms fragmented tracking and metric requests into a controlled, auditable workflow by layering persistent memory, skill deposition, unified gateways, and structured tool interfaces, ultimately delivering reusable rule assets and risk‑governed production hand‑offs.
The article analyzes the shortcomings of current AI‑assisted data‑warehouse development—context forgetting, unstable rule enforcement, and token‑heavy operations—and presents a five‑layer Harness architecture (persistent CLAUDE.md, Auto Memory, deterministic hooks, subagents, and SKILL refactoring) that systematically resolves these issues, boosts reliability, and embeds AI into the development pipeline.
The article explains how BP Claw tackles unstable AI coding results by automatically converting low‑quality PRD documents into structured, high‑quality requirements, applying token‑saving strategies, strict hallucination guards, and multi‑skill orchestration, which together boost FlinkSpec’s real‑time data‑warehouse delivery efficiency by up to 30%.
The article interviews two engineers who joined DeWu a year ago, revealing how structured mentorship, hands‑on project ownership, AI model optimization, and a supportive learning ecosystem accelerated their growth from beginners to contributors capable of leading complex technical initiatives.
