NLP Algorithm Expert

Responsibilities

Team Introduction: ByteDance's International Search Team is mainly responsible for the innovation of search algorithms for international short video products. We use the most cutting-edge machine learning technology for end-to-end modeling and continuous innovation and breakthroughs, while focusing on the construction and performance optimization of distributed systems and machine learning systems, from memory, Disk and other optimizations to the exploration of index compression, recall, sorting and other algorithms, to fully provide students with opportunities for self-growth. The main work directions include: 1. Including query analysis, relevance, the implementation of LLM technology in the field of multilingual search, pre-training technology and domain data SFT, etc., fully embracing large models, which is full of challenges 2. Cross-modal matching technology: Apply CV+NLP deep learning and multimodal fusion technology in search to give video search more powerful retrieval capabilities 3. Large-scale streaming machine learning technology: Apply large-scale machine learning to optimize the search experience and meet users' personalized needs 4. Architecture of hundreds of billions of data scale: From large-scale offline computing, performance of distributed systems, scheduling optimization, to building high-availability, high-throughput and low-latency online services, all aspects have in-depth research and innovation. 1. Lead the algorithm team: Responsible for the company's key international short video products, lead the algorithm team members, set team goals, and guide the team members in the development of search links, LLM and algorithm applications 2. LLM model application implementation: Responsible for the application of LLM in search business scenarios, optimize the search ranking mechanism and information efficiency, and explore multi-language pre-training, multi-modal fusion, etc., to explore the ideal form of search in the LLM era 3. Algorithm research and development: Participate in algorithm research and development work, improve model performance and algorithm efficiency, and continue to pay attention to the latest progress in academia 4. Project planning and management: Participate in project planning and management, formulate project plans, ensure that projects are delivered on time, and ensure high-quality completion of projects 5. Technology optimization and innovation: Continuously optimize existing algorithm technologies and promote algorithm innovation to improve business results and user experience 6. Cross-team cooperation: Work closely with product teams, engineering teams and business teams to understand business needs and convert algorithm technology into... is the actual product and solution 7. Algorithm evaluation and improvement: Responsible for evaluating and improving the algorithm model to improve the accuracy, efficiency and interpretability of the algorithm.

Qualifications

1. Master's degree or above, computer-related major, excellent schools in the field are preferred 2. Solid basic data structure and algorithm capabilities, code Solid ability 3. Familiar with the basic knowledge and common frameworks of machine learning and deep learning 4. More than 5 years of work experience, experience in NLP-related projects, familiar with common algorithms, and understanding of common NLP processing architectures 5. Have a good product concept , able to self-drivenly find and engineer solutions to product problems.