Large Recommendation Model Algorithm Engineer - Global E-Commerce

About the Team

The E-commerce Recommendation Foundation team focuses on developing next-level recommendation intelligence through a unified Foundation Model that supports diverse business needs and scenarios, encompassing the entire recommendation pipeline from initial retrieval to ranking and re-ranking. The team is pioneering an event-sequence-driven generative recommendation approach, deeply integrating technologies such as large language models (LLMs), multimodal perception, reinforcement learning, and system optimization to evolve recommendation systems into versatile intelligent agents.

We encourage original thinking, blending research with engineering craftsmanship, enabling team members to propose and validate new ideas openly. Your contributions, whether code or research, have potential to shape the future of recommendation technology. We seek individuals with broad intelligence perspectives to help redefine recommendation paradigms.

Key Responsibilities

• Design and enhance cross-scenario shared Foundation Models to support unified modeling and efficient inference workflows.
• Drive the advancement of generative recommendation paradigms centered on event sequence data by integrating multimodal understanding and generative functionalities.
• Utilize LLM technologies across retrieval, ranking, and re-ranking phases; engage in model training, optimizing inference, and system-level design collaboration.
• Investigate the synergy of LLMs and visual language models (VLMs) with recommendation systems to create smart, adaptive recommenders.
• Pursue research on end-to-end generative recommendation techniques and system optimization to effectively balance performance efficiency and user satisfaction.

Minimum Qualifications

• Strong theoretical knowledge in machine learning, deep learning, or information retrieval principles.
• Proficiency in Python programming and familiarity with popular deep learning frameworks such as PyTorch.
• A passionate, self-motivated research orientation toward intelligent recommendation technologies.

Preferred Qualifications

• Hands-on experience in large-scale recommendation systems development or training substantial models, with significant accomplishments in a related technical area.
• Background in research or publications concerning large language models, multimodal learning, reinforcement learning, or generative recommendation methods.
• Knowledge of pre-training and fine-tuning processes for large language or Foundation models.

About the Company

Established in 2012, ByteDance aims to inspire creativity and enhance lives globally. Its diverse portfolio includes products such as TikTok, Lemon8, CapCut, Pico, and several popular platforms tailored to the Chinese market like Toutiao, Douyin, and Xigua, facilitating content creation and consumption worldwide.

Why Work Here

ByteDance fosters a culture focused on inspiring creativity, with innovative products enabling authentic self-expression and global connectivity. The company values curiosity, humility, and impact-driven work to achieve continuous innovation and meaningful outcomes, promoting an "Always Day 1" mindset. It offers a collaborative environment striving for growth and breakthroughs together.

Diversity and Inclusion

Committed to an inclusive workplace, ByteDance embraces diverse skills, experiences, and perspectives from across the globe. The company’s mission to enrich life is supported through celebrating varied voices and fostering an environment reflective of its broad community.