Official JD technology sharing platform. All the cutting‑edge JD tech, innovative insights, and open‑source solutions you’re looking for, all in one place.
The article recounts a JD retail engineer’s rapid growth through tackling low‑resource large‑model training, developing a margin‑based dynamic data selection method (DynaMS) that earned an ICLR paper, and sharing practical insights on aligning business needs with cutting‑edge AI research.
This article explains the importance of business monitoring, distinguishes technical and business metrics, outlines a step‑by‑step process for constructing a business indicator system, and provides practical methods, tools, and common pitfalls for effective operations monitoring.
This article explains how Caffeine implements element expiration using three built‑in policies and a custom TimeWheel strategy, detailing the construction of the timer wheel, bucket selection, scheduling, advancing, and eviction logic within the cache's maintenance cycle.
This article provides a comprehensive walkthrough of Caffeine cache's architecture, explaining its fixed-size eviction policy, underlying data structures such as ConcurrentHashMap, MPSC buffers, Count‑Min Sketch frequency tracking, and the dynamic window‑probation‑protected zones, while detailing key methods like put, getIfPresent, and maintenance.
JD’s Merchant Intelligent Assistant leverages a large‑language‑model‑based multi‑agent architecture to provide 24/7 e‑commerce support, detailing its evolution, planning techniques, online inference, evaluation methods, sample generation, and practical insights for scalable AI‑driven operations.
This article presents JD Advertising's engineering experience with Apache Doris, describing the evolution from a data‑lake cold‑data solution to a native cold‑hot tiering approach, detailing performance regressions after upgrading to Doris 2.0, and outlining a series of optimizations for query speed, CPU and memory usage, schema‑change efficiency, and automated data migration and restoration.
Tech Insight presents concise overviews of ten JD retail technology papers accepted at top AI conferences in 2024, covering topics such as open‑vocabulary object detection, multi‑scenario ranking, diversity‑aware re‑ranking, a diversified product search dataset, semi‑supervised query classification, plug‑in CTR models, and methods to mitigate LLM hallucinations.
This article presents a complete MongoDB migration solution, detailing the migration rhythm, code refactoring using a decorator pattern, data source replacement with JImKV, MySQL and ES, bulk and incremental data transfer strategies, and deployment safeguards such as monitoring, gray release, and rollback to ensure a seamless cut‑over without service disruption.
This article details the correlation between system and business metrics, introduces three generic business‑monitoring platforms (UMP, PFinder, Taishan), defines a unified log format, provides Log4j and Java logging code, and explains alert rule configurations, visualizations, and real‑world incident case studies to improve operational reliability.
This article examines the challenges of outliers in large‑scale internet A/B testing, explains their statistical definition, outlines common causes, evaluates the benefits and limits of removal, and compares traditional trim and winsorize techniques along with practical detection and risk‑control strategies.
