Baidu Search Closed-Door Technical Symposium
The Baidu Search Closed‑Door Technical Symposium, the first core technical forum hosted by Baidu’s Search Architecture Department, brings senior engineers and junior backend developers together to discuss semantic retrieval, data‑driven big‑data processing, and vertical search offline architecture, while offering limited‑capacity sessions, networking gifts, and travel subsidies.
The Baidu Search Closed-Door Technical Symposium is the first core technical discussion forum hosted by Baidu Search, aimed at jointly exploring fundamental architecture technology issues with industry peers, exchanging ideas, and sharing Baidu's cutting-edge technology implementation practices.
The event is organized by the Baidu Search Architecture Department, which is responsible for the architecture design and implementation of Baidu's search system. It currently supports business areas including web search, video search, image search, multi-modal search, and e-commerce search, and builds high-performance, high-stability, elastic deployment, low-latency cloud-native retrieval systems, as well as high-throughput, low-cost offline systems for massive big data analysis and computing. Additionally, it applies and explores cutting-edge semantic technologies to continuously improve search effectiveness, enabling people to obtain information most equally and conveniently.
The symposium features three main topics, each hosted by Baidu T8-T9 senior engineers:
1. Semantic Retrieval Key Issues and Practices - Hosted by the Baidu Search Intelligence Architecture Leader. Topics include: How to integrate semantic recall technology with traditional search engines? How to optimize performance for semantic recall ANN technology? How to perform online and offline calculations for complex models at the ranking level? Common optimization techniques for heterogeneous computing?
2. Data-Driven Big Data Processing System - Hosted by the Baidu Search Data System Leader. Topics include: System cost - How to improve system cost-effectiveness by ten times? Data speed - How to improve average system data speed by ten times? Development efficiency - How to maximize system scalability and improve change efficiency by ten times; How to lower the development threshold and increase the number of parallel development personnel by ten times? Element synergy - How to achieve simultaneous significant improvement of all system capabilities, and what is the ultimate solution?
3. Baidu Vertical Search Offline Architecture - Hosted by the Baidu Vertical Search Offline Architecture Leader. Topics include: How to effectively crawl, process, and manufacture the vast and complex data on the internet, perform image visual feature calculations, build inverted indexes and vector clusters, and produce index data for user search. Discussion topics include: Storage of trillions of images and web pages on the internet; How to use distributed storage databases to support massive data storage and provide millions of QPS; For captured images, how to perform multi-dimensional image recognition, calculate image classification, pornography detection, clarity, OCR, disgusting content, portrait features, etc., involving distributed computing framework scheduling capabilities; How to effectively organize text and visual features for images to meet users' diverse search needs.
The event offers several benefits for attendees: 1) Participants who register and pass verification can randomly draw a small gift upon check-in, including dolls, ornaments, T-shirts, notebooks, etc.; 2) Smart Bluetooth earphones will be given to attendees who contribute the most to the content; 3) Transportation subsidies of 100 yuan JD.com cards per person will be distributed upon departure. To ensure attendee benefits, each topic is limited to 10 on-site participants, first come first served.
The event is recommended for backend engineers with 5 years or less experience. Registration is now open via the provided QR code or link.
Baidu Geek Talk
Follow us to discover more Baidu tech insights.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.