In the AI era, Qunar's front‑end tech director presents how D2C and large‑model‑based code generation dramatically improve front‑end productivity across three business scenarios, achieving up to 55% automated code coverage for both C‑end and B‑end applications.
This article describes how Qunar's technology center defined, measured, and managed system complexity through a custom modeling framework, implemented a dashboard for continuous monitoring, and established an anti‑corruption governance process that limits complexity growth to maintain low maintenance costs across hundreds of applications and systems.
This article describes Qunar's use of traffic recording and replay technology for fault isolation, performance optimization, and upgrade migration, detailing the evolution of their Java‑based agents, the end‑to‑end testing workflow, full‑link stress‑test integration, and the measurable business and performance outcomes.
This article examines the concept of business architecture, its origins in TOGAF, and presents a detailed case study of Qunar’s activity capability layer design, highlighting pain points, architectural solutions, and a step-by-step methodology for building robust, reusable activity systems.
In this interview, Qunar's technical director Zheng Jimin shares how the company revived Domain‑Driven Design to restructure hotel pricing, discusses strategic and tactical design, implementation challenges, success criteria, and the broader impact of technical influence on organizational growth.
This article compiles a selection of Qunar's 2021 technical writings covering architecture design, big data processing, front‑end engineering, and cloud‑native practices, providing titles, authors, brief abstracts, and direct links for readers seeking in‑depth engineering insights.
This article compiles the 2021 Qunar Technical Salon infrastructure series, presenting original technical writings on databases, operational practices, and core components, each linked to detailed posts that share real‑world experiences, design guidelines, and performance insights for engineers and practitioners.
In this interview, Zhu Shizhi, head of Qunar's infrastructure team, explains the evolution of the company's technical brand, the department's role in advancing cloud‑native and foundational technologies, key achievements such as open‑source collaborations and industry talks, and his vision for future technical branding efforts.
In this interview, Qunar's Technology Committee chair Wang Zhimeng shares how the company built its technology brand over the past two years, detailing the challenges, strategies, achievements, and future plans for fostering a strong technical community and enhancing talent attraction.
The article interviews Qunar's technology director to explore the current state of Domain-Driven Design in China, its business benefits and challenges, the company's practical experience with DDD and microservices, and provides guidance on how teams can quickly adopt DDD practices.
In 2020 Qunar's front‑end team explored multi‑platform adaptation, low‑code marketing, UI automation, security enhancements, and framework upgrades, publishing a series of technical articles that detail their solutions, performance gains, and practical insights for developers across the organization.
The September 7, 2020 technical salon hosted by Qunar's infrastructure R&D team, in collaboration with Keep, gathered over 20 engineering experts to discuss DevOps continuous delivery, test environment management platforms such as Noah, Portal, MPortal, and Thanos, and share practical experiences and architectural choices.
The article presents a series of interviews from Qunar's Technology Carnival, featuring experts Liang Zhangping, Wang Zhufeng, and Zheng Jimin who discuss hotel booking flow improvements, database architecture comparisons and migration to PXC, and comprehensive system stability governance practices.
The Qunar Technology Carnival features in‑depth interviews with experts Wang Mingyou, He Yang, and Jia Ziyan who share practical experiences on search ranking improvements, AIOps‑driven fault localization, and data‑driven revenue management, highlighting challenges, solutions, and future directions in AI‑powered systems.
This article summarizes Zhang Yan's presentation at the 2019 Gdevops Global Agile Operations Summit, detailing Qunar's OPS goals, evolution of its automation platform, the adoption of PHM concepts from aerospace to internet services, and practical fault‑prediction workflows, metrics, and challenges for achieving higher availability.
This article describes Qunar's comprehensive data security framework, covering data classification, warehouse access control, permission compliance, encryption, leakage detection, and the supporting institutional policies that together enable a small security team to protect large‑scale business data.
This article presents a case study of an overseas user experiencing video upload failures due to international roaming, explains how Qunar's network detection tool (QunarNDT) was used to diagnose the issue, outlines the findings about domestic routing and timeout, and offers recommendations for improving user experience.
The 2019 QInfrarch technical carnival in Suzhou gathered nearly 160 engineers from Qunar and partner companies to share in‑depth sessions on private cloud databases, real‑time hotel computation, AI‑driven ticket customer service, and OceanBase architecture, concluding with networking, tea breaks, and a lucky‑draw for attendees.
This article details Qunar's journey of automating application operations, covering the evolution of their host‑management system, unified monitoring/alert platform, and data‑interchange mechanisms that enabled the company to grow from a few hundred to over ten thousand servers with a stable six‑person ops team.
This article chronicles the five‑year evolution of Qunar's payment platform from a tightly coupled monolithic system to a high‑concurrency, highly available distributed architecture, detailing the motivations, service‑oriented splits, core components, and checkout variations that enabled support for diverse payment scenarios.
This article describes how Qunar replaced a low‑click hot‑words card with a personalized “You May Like” recommendation card, detailing data collection, rule‑based and collaborative‑filtering association methods, learning‑to‑rank models (subjective Bayes, RankBoost, LambdaMart), training‑sample strategies, online experiments, evaluation metrics, and future plans including LSTM‑based sequence modeling.
This article explains the basic principles of A/B testing, demonstrates a practical advertising experiment, describes effective experiment design, outlines Qunar's A/B testing platform architecture and workflow, and details statistical validation methods including Z‑test, minimum sample size calculation, and confidence interval estimation.
At Qunar, the data center algorithm team developed an automatic ticket classification system that combines Support Vector Machine with word2vec embeddings to handle high‑dimensional, low‑sample text data, achieving 89% accuracy and 80% recall while outlining the full machine‑learning pipeline from feature extraction to deployment.
