Machine Learning Engineer (LLM & Personalized Search), TikTok Search

Responsibilities

About The Team:
Our Search Team is responsible for building and owning TikTok's search engine which provides our users the best search experience. On the TikTok Search Team, you'll have the opportunity to build a full-stack search engine system and combine information retrieval technology with modern machine learning methods from related fields such as NLP, Computer Vision, Multimodal, and Recommender Systems. We embrace a culture of self-direction, intellectual curiosity, openness, and problem-solving.

About The Role:
You will be responsible for the personalization direction of photo-text search, lead the construction and implementation of the personalized algorithm system for photo-text search. Combine the core characteristics of the photo-text genre, focus on improving information acquisition efficiency and tapping the value of interactive comments, rely on large model technology to realize model scaling up and generative recommendation innovation, break through technical bottlenecks, expand the contribution of photo-text search to the search business, and create a user-centric search experience.

Key Responsibilities:
1. Be responsible for the implementation of personalized algorithms for photo-text search. Integrate photo-text visual, text and user behavior data to build scenario-adaptable personalized retrieval; rely on large model technology to promote model scaling up, optimize model performance and generalization ability, and explore the application of generative recommendation in photo-text scenarios to accurately match users' potential needs.
2. Focus on photo-text information acquisition efficiency. According to its characteristics of "visual priority, high information density and fragmented reading", reflect the differentiated value of photo-text and solve problems such as user interest mining through the personalized link.
3. Promote model iteration and effect review, collaborate with cross-departmental teams to decompose business objectives into algorithm strategies, drive photo-text search product innovation, and increase the contribution of photo-text in the overall search business.
4. Be responsible for the stability, performance optimization and cost control of the algorithm personalization link, promote full-link monitoring of the algorithm, improve response efficiency, and cooperate with the architecture team to enhance the training and inference efficiency of the personalization link to ensure stability and scalability.

Qualifications

Minimum Qualifications:
1. Bachelor's degree or above in a computer-related major, with excellent coding ability and a solid foundation in data structures and algorithms.
2. Familiar with and proficient in programming languages such as C++/Python, and familiar with common data structures and programming algorithms.
3. Solid foundation in machine learning and coding ability, with in-depth research experience in machine learning, NLP, CV, etc., and proficient in major algorithms and data structures.
4. Strong algorithm design and excellent ability to analyze and solve problems, passionate about solving challenging problems, good initiative and curiosity, and good communication and collaboration skills.

Preferred Qualifications:
1. Experienced in recommendation systems, advertising, search engines and other related fields.
2. Priority is given to candidates with R&D or implementation experience in model scaling up and generative recommendation
3. Strong technical innovation and problem-solving abilities
4. Strong business sensitivity and data thinking, able to identify optimization directions through data analysis, and possess good review and iteration abilities.