ML Engineer - Recommender Systems (Live Commerce)

TikTok; Seattle (WA); On-site; USD 154,000 - 301,000; Full time

Job summary

TikTok in Seattle is seeking an accomplished ML scientist at the PhD level to accelerate our next-generation recommender research. You will build and optimize models across recall, pre‑ranking, ranking, and multimodal signals to improve GMV, watch time, and long‑term user value.

You will partner with cross‑functional teams to turn research into scalable, measurable business impact, exploring generative recommendations, large models, RL, and long‑term value optimization.

Qualifications

• PhD or near-completion in a relevant field.
• Strong ML foundation with experience in recommendation systems or multimodal learning.
• Proficient in Python or C++, with PyTorch experience.
• Ability to translate business problems into effective modeling solutions.

Responsibilities

• Build and optimize models across recall, pre-ranking, ranking, and mixed ranking to improve GMV, conversion, watch time, and long-term user value.
• Develop cross-domain and multimodal modeling solutions that connect videos, live streams, products, and user behavior to power live commerce recommendations.
• Advance next-generation recommendation technologies, including generative recommendation, large models, reinforcement learning, and long-term value optimization.
• Partner with cross-functional teams to launch scalable solutions, run experiments, and translate research into measurable business impact.

Skills

PhD in CS/related

ML fundamentals

Python

C++

PyTorch

Data structures & algorithms

Analytical thinking

Education

PhD degree in CS/EE/Math/Stats or related

Tools

PyTorch

NumPy

CUDA

Job description

TikTok in Seattle is seeking an accomplished ML scientist at the PhD level to accelerate our next-generation recommender research. You will build and optimize models across recall, pre‑ranking, ranking, and multimodal signals to improve GMV, watch time, and long‑term user value.

You will partner with cross‑functional teams to turn research into scalable, measurable business impact, exploring generative recommendations, large models, RL, and long‑term value optimization.