Live Commerce ML Engineer: End-to-End Recommender Stack

TikTok Inc.; Seattle (WA); On-site; USD 207,000 - 368,000; Full time

Benefits offered by this job

Medical coverage (day one)

Dental coverage (day one)

Vision coverage (day one)

401(k) with company match

Paid parental leave

Disability coverage

Life insurance

Wellbeing benefits

10 paid holidays per year

10 paid sick days per year

17 days Paid Personal Time

Job summary

TikTok in Seattle is hiring to build the recommendation engine behind TikTok Shop LIVE. You will design and ship an end-to-end ML stack for live commerce with strict latency and compute constraints.

The role spans large-scale modeling, multimodal signals, and RL-style post-training, with ownership from data pipelines to online experiments and launch. Onsite work, competitive pay, and comprehensive benefits are provided.

Qualifications

• 3+ years of industry experience in large-scale ML, with models shipped and impact on online metrics.
• Strong fundamentals in retrieval/ranking, sequential modeling, multi-task learning, or generative/multimodal modeling.
• Proficiency in Python and/or C++, with PyTorch or TensorFlow and distributed training experience.

Responsibilities

• Scale Large Recommendation Models (LRM) by expanding foundation model capacity with sparse MoE and long user sequences.
• Translate offline scaling gains into online wins within latency and compute budgets.
• Build generative recommendations using multimodal Semantic IDs and near-real-time retrieval.
• Apply LLM4Rec and multimodal LLM modeling to unify live rooms, videos, and products.
• Develop reward models and post-training for long-term value and user experience.
• Design Agent x Recommendation systems to auto-tune multi-objective value weights and run experiments.
• Optimize for long-term value and ecosystem health while balancing GMV and user experience.
• Own end-to-end work from framing to data pipelines, training, serving, A/B tests, and launch.

Education

Bachelor's degree in Computer Science, Electrical Engineering, Mathematics or Statistics

Tools

PyTorch

TensorFlow

Python

C++

Job description

TikTok in Seattle is hiring to build the recommendation engine behind TikTok Shop LIVE. You will design and ship an end-to-end ML stack for live commerce with strict latency and compute constraints.

The role spans large-scale modeling, multimodal signals, and RL-style post-training, with ownership from data pipelines to online experiments and launch. Onsite work, competitive pay, and comprehensive benefits are provided.