About the team
The TikTok Search team is responsible for delivering a world-class search experience across TikTok's global ecosystem. Our mission is to connect users with the most relevant and high-quality content through advanced technologies in search ranking, recommendation, and multimodal understanding. As a campus hire, you will collaborate with top engineers and researchers to solve challenging problems at scale, working across the full stack of search—from content understanding and retrieval to ranking and user experience optimization.
Project Overview, Challenges & Value
With the rapid evolution of pre-trained large models, TikTok Search is advancing toward a new paradigm of AI-native search. Traditional search systems face increasing limitations when handling massive-scale data, multimodal content (video, text, image), and complex user intents. This project focuses on building a next-generation generative search system powered by large models to significantly enhance search intelligence and user experience.
Key focus areas include:
- Integrating large models with ranking systems to improve personalization and relevance.
- Developing end-to-end generative search models based on multimodal pretraining.
- Exploring large model–driven agent frameworks to better support complex, ambiguous, and multi-turn search scenarios.
Key challenges include:
- Personalized ranking: Fully leveraging multimodal signals beyond traditional model capacity.
- Ultra-large-scale retrieval and ranking: Scaling search systems to efficiently handle hundreds of billions of candidates.
- Complex query understanding: Accurately interpreting long, ambiguous, and multi-turn queries to improve user satisfaction.
Project Value:
- Technical Value: Drive innovation in large model–powered search architecture and enable agent-based AI search systems.
- Business Value: Significantly improve search relevance and user satisfaction, strengthening TikTok's search engagement and long-term growth.
We are looking for talented individuals to join our team in 2027. As a graduate, you will get opportunities to pursue bold ideas, tackle complex challenges, and unlock limitless growth. Launch your career where inspiration is infinite at our Company.
Successful candidates must be able to commit to an onboarding date by end of year 2027. Please state your availability and graduation date clearly in your resume.
Responsibilities
- Develop next-generation AI search systems using large models (LLMs).
- Improve ranking, personalization, and relevance in search.
- Build generative and multimodal search models (video, text, image).
- Explore LLM-based agents for complex and multi-turn queries.
- Optimize large-scale retrieval and ranking performance.
- Collaborate cross-functionally to ship research into production.Minimum Qualifications
- Individuals who are completing or recently completed a PhD in Software Development, Computer Science, Computer Engineering, or a related technical discipline.
- Solid foundation in machine learning, deep learning, or information retrieval.
- Experience in at least one of the following areas: LLMs, search/recommendation systems, ranking, NLP, or multimodal learning.
- Strong programming skills in Python and familiarity with ML frameworks (e.g., PyTorch, TensorFlow).
- Demonstrated research ability through publications, projects, or internships.
- Strong problem-solving skills and ability to work in a fast-paced, collaborative environment.
Preferred Qualifications
- Track record of research contributions through publications, open-source projects, or impactful research work.
- Hands-on experience with large-scale systems, distributed training, or real-world ML deployment.
- Experience with search, recommendation, or ads ranking systems in industry or research.
- Experience with multimodal models (video, image, text) or generative AI applications.
- Familiarity with LLM fine-tuning, alignment (e.g., RLHF), or agent-based systems.
- Strong interest in building end-to-end AI systems from modeling to production.
- Internship or research experience at leading technology companies or research organizations.
- Modeling experience in one or more of the following areas: Ads, Search engine, Recommender System, NLP/CV.
- Have a solid foundation in algorithms related to LLMs, including but not limited to comprehensive learning and practical experience in areas such as single-modal/multi-modal LLM application and deployment.
- Strong publications record in accredited conferences or journals（e.g., ICLR, NeurIPS, ICML, ACL, EMNLP, NAACL, CVPR, ICCV and ECCV）.