<p>About the Team The Search team is responsible for the machine learning algorithm for TikTok's rapidly growing video search, ecommerce search, local service search and AI search business. We use state-of-the-art large-scale machine learning technology, cutting-edge NLP, CV and multi-modal technology and generative models (LLM, VLM) to build the industry's search engines to provide the best search experience, for more than 1 billion monthly TikTok users around the world.</p> <p>About the Role TikTok is seeking an independent developer for the Search team. This role is pivotal in advancing TikTok's search capabilities by integrating cutting-edge recommendation algorithms and large language models (LLMs) technologies to deliver highly personalized and engaging user experiences.</p> <p>Responsibilities</p> <ul> <li>Independent execution: Independent development, experiment, analysis and deployment of large-scale personalized search and recommendation systems, ensuring scalability, efficiency, and robustness. </li><li>LLM Integration: Research and integrate LLMs within the recommendation pipeline to enhance content understanding, user intent recognition, personalization and content generation. Explore hybrid models that combine LLMs with traditional recommendation systems to mitigate feedback loops and uncover novel user interests. </li><li>Innovation with Generative Recommendation Systems: Drive the integration of generative recommendation approaches, leveraging generative models to directly generate personalized content recommendations, moving beyond traditional ranking-based methods. This includes exploring frameworks like GenRec, which utilize LLMs/VLMs to interpret user contexts and generate relevant recommendations. </li><li>Cross-functional Collaboration: Work closely with product managers, data scientists, infrastructure engineers, operation teams to enable search personalization strategies with overall product goals.Minimum Qualifications: </li><li>Educational Background: Bachelor's or advanced degree in Computer Science, Machine Learning, or a related field. </li><li>Technical Expertise: Proficient in machine learning frameworks (e.g., TensorFlow, PyTorch), programming languages (e.g., Python, Java, C++), and deep understanding of data structures and algorithms. </li><li>Industry experience: 2+ years in developing large-scale search or recommendation systems (feed, search, advertisement, etc). Familiarity with generative recommendation frameworks and their application in large-scale systems is highly desirable. </li><li>LLM experience: Hands-on experience with deploying and improving LLMs for real-world applications, including techniques like prompt engineering, supervised fine tuning, reinforcement learning, and retrieval-augmented generation. </li></ul> <p>Preferred Qualifications</p> <ul> <li>Global Experience: Experience working with international teams and understanding of diverse user behaviors across different regions. </li><li>Open Source Contributions: Active participation in open-source projects related to search, recommendation, or NLP. </li><li>Publications: Contributions to conferences such as NeurIPS, ICML, ACL, or RecSys are a plus. </li></ul>