Official platform of JD Retail Technology, delivering insightful R&D news and a deep look into the lives and work of technologists.
This article presents JD Retail's data lake architecture overhaul, detailing the shortcomings of the Lambda model, the migration to Flink‑Hudi‑Spark pipelines, performance gains, storage savings, unified APIs, and upcoming improvements for resilience and automation.
The article introduces AIGCDesign, an open‑source cross‑platform AI component solution that combines traditional frontend libraries with large‑language‑model capabilities, outlines its design goals, technical architecture, lifecycle hooks, configuration examples, multi‑framework support, and real‑world business integration cases.
The article narrates the author’s 30‑year programming career and shares technical reflections on building business‑agnostic, configurable data platforms, covering batch, streaming, interactive computing, big‑data sharding, Spark, Flink, cloud migration, and the philosophy of software architecture.
This article chronicles the multi‑generation architectural evolution of JD's advertising platform, explaining why a good code architecture must be extensible, reusable, maintainable and stable, and describing how domain‑driven design, aggregates, and ability orchestration were introduced to solve the challenges of complex, multi‑scenario business requirements.
The article details JD's adoption of the Taro cross‑platform framework to build a HarmonyOS native app, describing the technical architecture, CSS and rendering optimizations, performance gains over native ArkTS, and future plans for further speed and flexibility improvements.
This article recounts the development of JD.Vision as a native Apple Vision Pro application, detailing how spatial computing reshapes e‑commerce experiences, the challenges of home‑scene product visualization, user‑type strategies, and the product‑manager growth lessons drawn from the project.
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.
This article presents the design goals, architecture, technology selection, and component details of a unified video review system that leverages FFmpeg for frame extraction, stream‑based parallel processing, and flexible synchronous/asynchronous workflows to achieve low latency and high scalability.
The article details JD Advertising's technical challenges and solutions for large‑scale sparse recommendation models, describing GPU‑focused storage, compute and I/O optimizations for both training and low‑latency inference, including distributed pipelines, heterogeneous deployment, batch aggregation, multi‑stream execution, and compiler extensions.
This article examines JD Retail's content compliance platform, detailing user‑facing problems, business‑level audit responsibilities, key performance metrics, operational workflows, and a technical case study on detecting price over‑pricing using comparable‑price models and large‑scale price prediction.
