Machine Learning Engineer, Commerce Ads Ranking

This on-site role focuses on optimizing models across the TikTok Commerce Ads Ranking pipeline, from recall through final ranking, and on algorithm and system co-design plus large-scale distributed training within privacy-preserving environments. The position is based in San Jose, California.

Responsibilities

• Optimizing models in the ads delivery system, including feature engineering, model architecture, auto crossing, ads cold start handling, modeling delayed feedback, multi-task learning, and sequence modeling.
• Algorithm and system co-design, covering retrieval algorithms, sample mining, and long sequence processing.
• Exploring large-scale distributed training frameworks with focus on GPU tuning, feature processing, and synchronous training to enhance ads delivery efficiency in privacy-preserving environments.
• LRM, the next generation recommender system leveraging an LLM learning paradigm: entity understanding, end-to-end Generative Recommendation, sequence-only Recommendation, and a Mixture of Export approach.

Requirements

• Strong coding skills in C++ and Python, with a solid foundation in data structures and algorithms.
• Solid foundation in machine learning, familiarity with common algorithmic models such as FM, LR, GBDT, DNN, Wide & Deep, etc.
• Excellent learning ability and good teamwork spirit, with strong analytical and problem-solving skills. Publications in conferences such as KDD, NeurIPS, SIGIR, etc., or experience in data mining or machine learning competitions are a good fit.
• Bachelor’s degree or higher in computer science or related majors.

Technologies

• C++
• Python
• LLM

Benefits

• Medical, dental, and vision insurance from day one
• 401(k) savings plan with company match
• Paid parental leave
• Short-term and long-term disability coverage
• Life insurance
• Wellbeing benefits
• 10 paid holidays per year
• 10 paid sick days per year
• 17 days of Paid Personal Time, prorated at hire with accruals increasing with tenure
• Eligible for discretionary bonuses or incentives
• Restricted stock units

Los Angeles County (Unincorporated) Candidates

• Interacting and occasionally having unsupervised contact with internal and external clients or colleagues
• Appropriately handling and managing confidential information, including proprietary and trade secret information and access to information technology systems
• Exercising sound judgment