Responsibilities
Team introduction: The international short video search team is mainly responsible for the search algorithm innovation and architecture research and development of international short videos. We use the most cutting-edge machine learning technology for end-to-end modeling and continue to innovate and make breakthroughs. At the same time, we focus on the construction and performance optimization of distributed systems and machine learning systems, from memory and disk optimization to the exploration of index compression, recall, sorting and other algorithms, fully providing students with opportunities to grow themselves. The main work directions include: 1. Exploring the most cutting-edge NLP technology: from basic word segmentation, NER, to application Query analysis, basic correlation, etc., applying deep learning models in the entire link, every detail is full of challenges 2. Cross-modal matching technology: applying CV+NLP deep learning technology in search to give video search a more powerful retrieval capability 3. Large-scale streaming machine learning technology: applying large-scale machine learning to solve the recommendation problem in search, making the search more personalized and better about you 4. Architecture for hundreds of billions of data scale: There is in-depth research and innovation in all aspects from large-scale offline computing, distributed system performance and scheduling optimization to building high-availability, high-throughput and low-latency online services. Mainly responsible for: 1. Participating in the improvement of search recommendation models and strategies for international short videos, as well as international short e-commerce, life services and other key businesses, and responsible for the search traffic and user penetration growth of these businesses & search mind building tasks 2. Using recommendation algorithms as the core technology stack, improving the recommendation system based on ultra-large-scale machine learning models, covering the full-link technology links from candidate mining to recall, rough ranking, fine ranking, multi-objective fusion 3. Exploring short text recommendation and general... The upper limit of recommendation technology focuses on the joint application of recommendation and NLP technology, as well as the exploration of cutting-edge technologies such as multi-modality. Business introduction 1. Search growth business: The functions and scenarios the team is responsible for basically cover the vast majority of search traffic and were the biggest reason for the growth of TikTok search traffic in the past. The means include guiding/stimulating/facilitating the entire process before the search occurs, such as recommended query scenarios inspired by videos/comments in the main feed, input completion before search, and result-related search scenarios after search. It not only brings more traffic to search, but also makes the unit value of the traffic itself higher 2. E-commerce search growth business: E-commerce is an important monetization method for the app, and search is a key part of the construction of shelf mentality. The growth of e-commerce search traffic and the establishment of mentality play an important role in it 3. The combination of search and terminal: As a search business, it is also responsible for the search-related sorting logic in FYP ranking, changing the ecology of the terminal to stimulate users willingness to search and explore more content. And use users search behavior to provide users with a better feed browsing experience.
Qualifications
1. Basic requirements: good algorithm design capabilities and engineering implementation capabilities, and practical experience in machine learning/reinforcement learning/NLP applications (Part 1) 2. Have good communication and expression skills, and have a clear understanding of good user experience. Candidates with good product awareness are preferred. Bonus points: 1. Applicants with experience in optimizing large-scale (large candidates, large number of users) recommendation systems are preferred 2. Applicants with experience in pre-training large-scale NLP language models are preferred 3. Applicants who have participated in competitions such as ACM and achieved good rankings are preferred.
