The article distinguishes between cost cutting (reducing bills) and cost avoidance (slowing bill growth relative to business), explains why line-level optimizations rarely affect financial statements, and argues that real 30% savings come from platform-level levers like oversubscription ratios, depreciation policies, and billing strategies, requiring centralized management and end-to-end efficiency metrics.
JD presents its Agentic advertising paradigm where AI agents replace human decisions across advertiser, consumer, and platform layers, detailing the Jing Xiaotong B2B agent, a reasoning-based delivery brain, conversational C2C shopping agents, and a super-node compute infrastructure achieving 90% latency reduction and 300TB cache for millisecond-scale inference.
The article analyzes JD.com's new AI payment interface, showing how AI-initiated payments compress the entire transaction-to-payment loop into a single voice command, shifting identity verification from human to AI, making payments seamless and invisible compared to traditional quick payment methods.
At JDD 2026, JD Advertising's Zhang Zehua details how AI agents are replacing human decision-makers across advertiser, consumer, and platform layers, showcasing the Jing Xiaotong autonomous advertising agent, a reasoning model boosting ROI 8.7%, and conversational shopping agents handling 70% vague queries via TTR and RIGER architectures.
AI agentsAdvertising TechnologyAgentic Advertising
At its 2026 global tech conference, JD.com revealed a 100,000-card domestic compute cluster, the JoyAI-Echo world model scoring 81.6 on WBench Navigation, 10 million hours of embodied data collection, and a Physical AI Acceleration Plan targeting six global firsts in robotics data, manufacturing, parts supply, retail, deployment, and maintenance.
JD’s Oxygen AIIC combines human‑in‑the‑loop ontology engineering, a semantic search‑then‑discrimination pipeline, and a self‑evolving multi‑task LLM/VLM model to produce high‑quality product knowledge for over a hundred thousand categories and billions of daily SKU updates, boosting search coverage to 80%, attribute auto‑fill to over 80%, cutting quality issues by 37% and raising click‑through by 9% while achieving 94.2% precision and 82.8% recall.
This tutorial shows how to build a JD.com search URL, fetch the page with urllib, parse the HTML using BeautifulSoup, and reliably extract each product's name, link, image and price while handling missing image URLs and avoiding regex complexity.
This tutorial shows how to build a JD.com search URL, encode keywords, fetch the page with Python's urllib, and extract product details using regular expressions, providing code snippets, regex explanations, and sample output for beginners.
JD’s recent technical salon reveals the rapid evolution of large‑model tools, detailing industry trends, JD’s JoyAI ecosystem—including JoyAgent, OxyGent and JoyCode—real‑world applications across office, code review, logistics and local services, and future policy and multi‑agent visions.
JD’s internal technical salon reveals the rapid evolution of large‑model AI tools, outlining design principles, current capabilities like JoyAgent, OxyGent and JoyCode, real‑world applications across office, code review, logistics, and future policy and multi‑agent visions, highlighting their impact on modern workplace productivity.
This tutorial explains how to use Python to scrape product reviews from JD.com via its AJAX comment API, extract fields such as nickname, score, content, and image count, and save the collected data into a CSV file using the requests and csv modules.
The article presents a comprehensive adaptive degradation and automatic recovery framework for JD Alliance's recommendation system, designed to handle high‑frequency, instantaneous traffic surges during large promotions by combining real‑time monitoring, Wilson‑interval‑based timeout correction, scenario‑aware control, traffic slicing, linear‑programming‑driven chain optimization, and low‑cost business‑agnostic APIs, achieving over 90% reduction in traffic loss and zero incidents.
This article details JD.com's challenges and solutions for deploying domestic NPU chips across heterogeneous GPU‑NPU clusters, covering architecture, scheduling, high‑performance training and inference engines, real‑world case studies, and future plans to scale AI workloads securely and efficiently.
This article explores how JD.com's advertising team leverages Agent technology to enhance advertising operations through AI-driven automation, covering technical implementations like RAG, Function Call capabilities, and platform architecture for scalable AI solutions.
This article details how JD's BIP procurement system tackled massive query‑performance challenges by segmenting order data, leveraging the JED distributed MySQL solution, introducing JimKV for hot‑data caching, and offloading complex searches to Elasticsearch, resulting in dramatically reduced load and faster user experiences.
JD.com’s data‑governance framework combines a health‑score‑driven, automated platform that cross‑verifies audit logs, builds full‑link and operator‑level lineage, introduces standard fields, and optimizes resource mixing, task staggering, and cross‑datacenter scheduling, while targeting real‑time AI‑enhanced detection and full automation.
This article presents JD.com’s comprehensive big‑data governance experience, detailing the background and challenges, the automated governance platform and its core technologies such as audit logs and full‑link lineage, strategies for resource optimization, and the roadmap toward real‑time, intelligent, and fully automated data governance.
The author builds a fully AI‑driven pipeline that scrapes JD.com comments about an Elasticsearch book, processes the data through cleaning and preprocessing, indexes it into Elasticsearch, and creates a series of Kibana visualizations, while reflecting on model selection and practical challenges.
This article presents JD.com’s multimodal recommendation system architecture, covering content understanding, multimodal ranking and recall models, practical deployment pipelines, and future research directions such as large‑model integration and supply‑side generation, all illustrated with detailed diagrams and Q&A.
In this interview, JD Retail’s AI director Zhai Zhouwei outlines the four historical stages of e‑commerce search, explains how large‑model AI is reshaping user interaction, retrieval and content generation, discusses practical challenges and solutions, and shares his vision and advice for enterprises adopting these technologies.
This article describes how JD Alliance built an adaptive degradation and automatic recovery framework for its recommendation system to handle sudden, large‑scale traffic spikes during major sales events, ensuring stability while minimizing recommendation loss through real‑time monitoring, scenario‑aware control, and linear‑programming‑based pipeline orchestration.
The JD Group's Project Management Talent Development Channel presented a series of expert talks and a round‑table at the third China PMO & PM Conference, covering data‑driven project intelligence, best practices in internet project management, AI‑enhanced workflows, and overseas delivery challenges, while highlighting JD's commitment to cultivating project management excellence.
JD.com’s Project Management Talent Development Channel showcased its R&D efficiency, AI‑enhanced workflows, and industry‑specific project practices at the 3rd China PMO&PM Conference, offering insights from multiple experts on overseas delivery, healthcare projects, and systematic talent cultivation, while earning a special honor from the PM&PMO community.
On August 10‑11, JD Group's Project Management Talent Development Channel showcased its expertise at the 3rd China PMO&PM Conference in Shanghai, delivering multiple expert talks on data‑driven project management, R&D efficiency, AI‑enabled workflows, overseas delivery, and fostering industry dialogue through round‑table discussions.
This article outlines the JD Insurance UED team's recent explorations and achievements in leveraging AIGC technology to transform creative design workflows, highlighting key insights, visual examples, and recommended readings to inspire broader adoption of AI‑generated content tools across the industry.
This article presents an in‑depth overview of JD.com’s retail data visualization platform, detailing its product matrix—including EasyBI, a low‑code platform, and JDV large‑screen tool—its architectural layers, key capabilities, business case studies, challenges faced, and future development directions.
The article presents a comprehensive overview of JD's advertising retrieval platform, detailing how it balances limited compute resources with massive data through adaptive compute allocation, distributed execution graphs, elastic systems, and multi‑stage algorithmic improvements to achieve high‑performance, scalable ad matching.
This article chronicles the author's three‑year journey as an AI product manager at JD.com, detailing the decision to enter the AI field, key project experiences such as platform integration and edge‑intelligence, a three‑stage growth roadmap, essential capabilities across four quadrants, and a forward‑looking view of AI's industry impact.
JD.com became the first Asian company to win the prestigious INFORMS Prize by leveraging AI‑powered, user‑centric supply chain innovations that deliver 95% of orders within 24 hours, scale assets to 153.8 billion yuan, and generate tens of thousands of daily logistics decisions.
This article presents an in‑depth overview of JD's retail data visualization platform, detailing its product matrix (EasyBI, low‑code platform, JDV), real‑world business use cases, architectural challenges, future development strategies, and a Q&A session that highlights technical and operational insights.
This technical presentation describes JD's data service platform, covering its origin, performance optimizations, flexible API generation, scaling to massive metrics, caching strategies, service orchestration, governance, and a Q&A on security and data‑source flexibility.
The article recounts JD’s journey from a small team building an in‑house advertising system that saved billions, through successive PC, mobile, and middle‑platform eras, to current AI‑driven retail innovations, highlighting strategic decisions, architectural challenges, and the importance of aligning technology with business goals.
This article presents JD Retail's digital‑twin framework for its complex supply‑chain network, detailing the underlying AI‑driven intelligent diagnosis, end‑to‑end simulation platform, two concrete case studies on inventory sinking and multi‑warehouse layout optimization, and outlines future research challenges and opportunities.
This whitepaper outlines JD.com's practical zero‑trust security approach, detailing a five‑element framework that includes asset digitization, asset identity, diversified security checkpoints, a strategy center, and a zero‑trust cockpit, to help digital enterprises strengthen security, reduce costs, and meet regulatory requirements.
This article presents JD's super automation approach for its supply chain, detailing the business background, challenges, AI‑driven forecasting, procurement, intelligent allocation, inventory clearing, integrated decision making, and future directions toward fully automated, optimal end‑to‑end operations.
The article outlines JD's data‑driven business development strategy, describing the current challenges of its business data marketplace, the governance framework—including layered architecture, standardization, ClickHouse dictionary refresh, and optimization measures—and the resulting performance improvements and future outlook.
This article presents a comprehensive case study of how JD.com tackled the growing load of its shopping‑cart service by introducing an asynchronous pagination mechanism for auxiliary product information, detailing the background, objectives, technical architecture, challenges, and measurable performance and cost benefits.
This talk by JD Retail's Zhao Yu explains how reinforcement learning is modeled and deployed for large‑scale traffic control during major sales events, detailing system architecture, reward design, offline simulation, model upgrades, and future research directions.
On July 16, JD.com’s jointly developed intelligent product‑matching technology received the second‑place Science and Technology Progress Award at the 11th Wu Wenjun Artificial Intelligence Award, highlighting its multi‑dimensional user profiling, cross‑modal product modeling, and precise matching innovations that have driven significant commercial and societal impact.
This article presents a comprehensive technical overview of JD.com’s retail traffic data processing pipeline, detailing the multi‑layer data warehouse architecture, real‑time and offline data flows, a large‑scale back‑fill case using Iceberg and OLAP, data‑skew detection and mitigation techniques, and future directions involving unified Flink‑Spark streaming‑batch solutions.
This article introduces JD.com's ISV co‑development platform for the gold‑flow of its APP, detailing the background, overall solution design, implementation steps, capability panorama, co‑building workflow, quality control mechanisms, real‑world case studies, and future outlook to illustrate how the model accelerates business development and improves ROI.
The article details JD International's challenges and solutions in building a unified, real‑time data platform for its multi‑regional 618 promotion, covering business characteristics, data distribution, team organization, dashboard architecture, integration strategies, and short‑ and long‑term technical plans.
The article presents MCube, a cross‑platform native dynamic UI framework used in JD.com’s mobile app to improve performance, reduce development effort, and enable rapid, stable feature delivery through a modular architecture, XML DSL, Yoga layout engine, and comprehensive management and tooling support.
The article examines the evolution of project management standards, highlights China’s recent national guidelines, and presents JD.com’s intelligent shopping guide AI project as a case study illustrating how business‑driven value streams, advanced technologies and rigorous target‑benefit management can shape modern project management practices.
This article presents JD's comprehensive big‑data security framework and distributed permission system, detailing the overall planning of the security center, data lifecycle protection strategies, core modules such as subjects, resources, policy language, and high‑performance access control, and how they address national compliance, business scalability, and technical challenges.
This article traces the evolution of big data technologies from early concepts and Google research papers through the rise of Hadoop, examines the platform transformations of Alibaba, JD.com, and Didi, and offers practical stack‑selection guidance for medium‑ and small‑scale enterprises.
This article explains how JD.com designs and evolves its OLAP platform, covering data sources, ingestion, storage, real‑time and offline processing, key challenges such as timeliness, high throughput, consistency, and the solutions implemented to support massive e‑commerce analytics.
JD's OLAP platform runs on ClickHouse and Doris across 3,000 servers, handling billions of daily queries and petabytes of data, and this article details the selection criteria, cluster deployment models, high‑availability architecture, operational challenges, and future roadmap.
This tutorial walks you through using Python to collect JD.com product comments, parse the JSON responses, store the data in Excel, and then perform simple visual analysis such as color‑distribution bar charts and a word‑cloud of review content.
This article explains JD.com's end‑to‑end order fulfillment process, the concepts of 211/411 delivery promises, the mechanisms behind order splitting by warehouse, merchant and payment dimensions, and how monetary discounts are proportionally allocated across SKUs.
JD.com’s order center faced massive query loads, prompting a migration from MySQL to Elasticsearch and a multi-stage evolution of its ES cluster—from a chaotic initial setup to isolated nodes, replica tuning, master‑slave dual clusters, and real‑time backup—culminating in a robust, high‑throughput, fault‑tolerant search architecture.
This article introduces JD.com's 9N GRAPH industrialization framework for large‑scale graph algorithms in advertising, covering the challenges of e‑commerce recommendation, the end‑to‑end solution architecture, the BVSHG multi‑behavior heterogeneous GNN model, training pipelines, and observed business impact.
The article outlines JD’s design of a unified social‑relationship front‑end module, detailing current data‑island problems, user‑value requirements, visual guidelines, functional integration, and simplified friend‑adding flows to boost acquisition, engagement, and conversion across the platform.
In this interview, senior architect Li Zunjing shares his two‑stage journey at JD.com—from stabilizing e‑commerce systems to pioneering B‑PaaS—highlighting the importance of technical fundamentals, a "empty‑cup" mindset, collaborative execution, and self‑driven learning for engineering success.
This article explains the fundamental differences between traditional and cloud‑native monitoring systems, outlines the challenges each faces, and details JD.com's evolution from physical servers to JDOS 2.0, describing its modular architecture, deployment model, and ongoing optimization efforts.
This article explains how JD constructs its OLAP platform from data ingestion to storage, querying, and management, describing the diverse data sources, real‑time and offline processing, scalability, consistency, fault tolerance, and future optimization plans, while addressing key technical challenges and solutions.
This article describes how JD.com leverages the open‑source MPP analytical database Doris for real‑time and offline OLAP on customer‑service data, covering data ingestion pipelines, dual‑stream high‑availability design, dynamic partition management, multi‑level caching, monitoring with Prometheus‑Grafana, and performance optimizations applied during major sales events.
The article explores why enterprises must continuously innovate, outlines the challenges of cross‑industry disruption, digital transformation, and VUCA environments, and presents JD.com’s lean innovation process, S‑curve growth model, three‑horizon strategy, and concrete internal innovation cases to drive sustainable growth.
The article outlines JD.com’s strategic focus on integrating artificial intelligence, blockchain, cloud‑edge computing and other emerging technologies with the real economy to build a trustworthy, digital‑intelligent supply chain that drives high‑quality development and empowers millions of enterprises.
This guide walks you through installing the jd_seckill Python script, configuring JD.com credentials, setting the purchase time, and executing the tool to automate Moutai flash sales, including environment setup, required dependencies, and step‑by‑step usage instructions.
This article recounts the JD.com 11.11 tech salon presentation where architect Shi Weihang explains how the NeuHub gateway was redesigned to handle over two billion TTS service calls, detailing the challenges, scaling techniques, and architectural decisions that enabled such massive traffic.
This article provides a step‑by‑step guide on using Scrapy’s crawl template to automatically scrape product details such as ID, title, shop name, shop link, and price from JD.com, including source analysis, project setup, code snippets, and result verification.
This article summarizes Bao Yongjun’s presentation on JD.com’s end‑to‑end big data platform, covering its strategic value, industry trends, architectural design, development phases from scale‑out to intelligent real‑time processing, and future directions for a cloud‑native, AI‑driven data ecosystem.
The HotKey middleware, built on JD’s 618 stable version 0.2, introduces protobuf serialization and optimizes data transmission, achieving up to 37 k QPS stable throughput, 15% lower CPU peaks, and millisecond‑level hot‑key detection and push across dozens of core JD services, dramatically reducing backend storage load.
This article explains how federated learning addresses data‑island challenges under privacy regulations, introduces JD.com’s 9NFL federated learning platform, details its component‑based, high‑availability, high‑performance architecture, and showcases a successful advertising use case that boosted revenue by over 10%.
This event features an architect from JD.com's backend team discussing strategies for designing high-availability, high-performance order systems to handle massive concurrent orders during major sales events like 618, based on real-world case studies and Elastic technology implementations.
The 2020 JD.com 618 shopping festival showcased a record-breaking sales volume and demonstrated how meticulous resource fine‑tuning, container scaling, high‑fidelity load testing, and cross‑team coordination enabled stable, cost‑effective system performance without adding new compute servers.
This article details JD's comprehensive redesign of the Hundred Billion Subsidy marketing brand, covering project background, identified pain points, the creation of a distinctive IP, visual positioning, brand distribution standards, and data‑driven UI improvements that boosted user attraction, coupon redemption, and overall performance.
This article details how JD.com’s retail teams collaboratively designed, built, and launched the lightweight Jingdong Fast Version mobile app in early 2020, outlining the project’s tight timeline, cross‑team coordination, technical architecture, design simplifications, and the successful on‑time release despite pandemic constraints.
This case study examines JD.com’s evolution into a technology‑driven enterprise, detailing its corporate culture, the “ABCDE” technology strategy, the implementation of DevOps and agile practices through the CALMS framework, and how unified continuous‑delivery platforms and operational metrics have driven growth, efficiency, and pandemic response.
This article details how JD.com’s order center migrated its Elasticsearch cluster through multiple architectural stages—initial deployment, isolation, replica tuning, master‑slave adjustments, and real‑time dual‑cluster backup—while addressing data synchronization, scaling, and performance pitfalls to achieve high availability and query stability.
The article explains how focusing on a single critical point—such as the deployment stage—can dramatically accelerate an organization’s end‑to‑end DevOps transformation, illustrated with JD.com’s journey from manual releases to an automated, high‑efficiency continuous delivery platform.
The article summarizes Zhu Qingjie’s presentation on JD.com’s core transaction platform, highlighting its massive scale, stringent high‑concurrency and high‑availability requirements, architectural strategies such as system splitting, caching and componentization, and the use of an autonomous decision platform to ensure rapid, reliable e‑commerce operations.
Despite a 10.3% sales decline in the broader auto market, JD.com’s automotive unit achieved explosive growth by combining ToC and ToB services, building a comprehensive parts database, linking online and offline channels, and breaking information silos with a robust middle‑platform strategy.
This article details how JD.com’s order center migrated its Elasticsearch cluster from a basic, mixed‑cloud deployment to a real‑time dual‑cluster backup solution, covering each architectural stage, scaling decisions, data‑sync strategies, and the performance pitfalls encountered along the way.
JD Retail’s technology and data platform adopted agile methods, launched training and coaching, created an Agile Project Management Guide, and achieved a 25% reduction in delivery cycles while 80% of projects now follow agile practices, showcasing measurable efficiency gains.
JD.com's 2019 JDDiscovery conference revealed a comprehensive, cloud‑native technology landscape that spans AI, big data, IoT, and blockchain, detailing how the company has transformed its integrated retail, logistics, and finance systems into modular, open‑service solutions for external partners.
The article details JD.com's coordinated effort to prepare for the 11.11 shopping festival by expanding load‑testing scale, improving cross‑team collaboration, and enhancing the ForceBot platform with traffic recording and replay capabilities to achieve more realistic and efficient full‑chain performance evaluations.
In this talk, JD Retail's Liu Guangyue shares how product managers should adopt a higher‑level, ecosystem perspective to design mid‑platform products, reflecting on current component‑based e‑commerce platforms, past social‑product challenges, and future strategic product design principles.
The article chronicles JD engineer Zheng Ruifeng’s path to becoming a Spark Committer, highlighting his early involvement, key contributions to Spark’s ML and GraphX components, the community’s scale, and his vision for future improvements in the big‑data platform.
This article demonstrates how to use Java HttpClient to call the free JD.com Wanxiang news channel API, providing a fully commented implementation, response handling methods, and a sample JSON output for developers who need a clear, unwrapped example.
The article outlines JD’s evolution from rule‑based product suggestions in 2012 to a sophisticated, AI‑driven, multi‑screen personalized recommendation platform, detailing its product types, system architecture, data collection, offline and online computation, and the core recommendation engine that powers features like “Guess You Like.”
The article details JD.com’s journey from rule‑based product recommendations in 2012 to a sophisticated, AI‑driven personalized recommendation system, describing its multi‑screen product types, data collection, offline and online computation pipelines, and the modular architecture of its recommendation engine.
This article demonstrates building a Python web scraper to collect JD.com product reviews, extract product IDs, store comments in MongoDB, clean color and size data, and visualize the most popular colors and sizes using matplotlib, providing a complete end‑to‑end data analysis pipeline.
This article examines JD.com's massive HBase deployment, detailing its evolution from early adoption to a 90PB, 7,000‑node cluster, the platform's architecture, multi‑active disaster recovery, multi‑tenant isolation, and the integration of Phoenix for SQL‑based access, offering practical insights for large‑scale distributed storage.
The JDATA Absolute Semantic Recognition Challenge concluded on July 15 with eight finalist teams presenting their models, the champion "Label Team" using BERT, judges highlighting data cleaning, model fusion, and the distinction between absolute and contextual words, and a detailed ranking of all participating groups.
This article presents a technical case study of JD.com's 6.18 promotion, detailing an AI‑driven intelligent text recognition system for address parsing and a high‑performance order‑trajectory visualization pipeline that combines multiple algorithms to achieve high accuracy and 99.99% availability.
JD.com’s R&D team transformed its 618 and 11.11 promotion preparation from a six‑month sprint to a one‑to‑two‑month process by building a unified middle platform that combines component‑based technology and data services, enabling rapid, scalable, and cost‑effective retail operations across multiple scenarios.
This article explains how JD.com leverages the open‑source HBase database for massive, low‑latency data storage across various business lines, detailing its architecture, multi‑tenant isolation, disaster‑recovery mechanisms, and integration with Phoenix SQL for OLTP workloads.
This article introduces Apache ShardingSphere’s ecosystem, core sharding and governance capabilities, access endpoints such as Sharding-JDBC, Sharding-Proxy and Sharding-Sidecar, and details JD.com’s real‑world implementation, including data‑sharding strategies, distributed primary keys, hint routing, performance optimizations, and future roadmap.
The article chronicles JD.com’s turbulent CTO history, detailing past security incidents, major system overhauls, and strategic shifts toward cloud, big data, and AI, while highlighting how leadership changes reflect the company’s push to become a technology‑first enterprise.
The article presents a detailed technical interpretation of JD’s Design Help Open Platform, describing how it aggregates designers, leverages 3D modeling, VR, and fast rendering to solve the pain points of home‑decoration users, improve designer efficiency, and create a closed‑loop shopping experience.
JD.com’s overseas data analysis platform, dubbed the Columbus platform, combines a lightweight data warehouse deployment with standardized, customizable BI tools to provide real‑time and offline analytics, visualization, KPI management, and future self‑service reporting and predictive capabilities for its global e‑commerce operations.
This article details how JD.com’s order center migrated its Elasticsearch cluster through multiple stages—from an initial unoptimized deployment to a real‑time dual‑cluster backup solution—addressing scalability, reliability, shard tuning, version upgrades, and data synchronization strategies to support billions of documents and hundreds of millions of daily queries.
JD.com's intelligent retail technology, which integrates online and offline retail through digital technology, has won a special award for technological industrialization in intelligent retail innovation, marking the first time an e-commerce company has received this honor.
At the INFORMS annual meeting in Phoenix, JD.com’s chief supply‑chain scientist and a team of data scientists presented a series of talks on big‑data‑driven supply‑chain optimization, demonstrated cutting‑edge logistics technologies, and engaged with global scholars and industry leaders, highlighting JD’s research strength in operations and analytics.
The article details JD.com’s 3.5‑generation delivery robots, explaining their high‑precision multi‑sensor localization, deep‑learning perception, reinforcement‑learning control, extensive patent portfolio, and future challenges, while also inviting readers to vote for the robots in a national patent competition.
In this interview, senior JD.com project manager Tian Yanyun discusses his most memorable projects, coordination techniques without formal authority, demand decomposition, feasibility assessment, risk control, portfolio benefits, and how to thrive in a volatile, uncertain, complex, and ambiguous environment.
Fibonacci is JD.com's enterprise-grade serverless function platform combining container technology, serverless architecture, and event-driven mechanisms to provide high-availability, high-efficiency, and high-usability FaaS solutions.