About the job
Our Foundations and Intelligence Service R&D team is fast growing and responsible for building state-of-the-art foundation models, such as LLM, VLM and Omni Models. Our mission is to build a bridge for collaboration between foundation models and downstream business scenarios, and use foundation model powered world knowledge to enhance better user experiences across TikTok, including content moderation, search and recommendations, client AI, etc.
Responsibilities
- Lead the incubation of next-generation, high-capacity LLM solutions for TikTok business, identify and define both short and medium term objectives;
- Design methods, tools, data recipes and experiments to push forward state-of-art in large language models;
- Explore new model architecture and inference-efficient model design for LLM applications to scale impact on business
- Work closely with cross-functional teams to plan and implement projects harnessing LLMs for diverse purposes and vertical domains
- Extend the insights and impact from industry to academia
Qualifications
Minimum
- Ph.D in Computer Science, Data Science, Artificial Intelligence, or a related field
- Proficiency in programming languages such as Python, Rust, or C++ and a track record of working with deep learning frameworks (e.g., pytorch, deepspeed, megatron, vllm, etc.)
Preferred
- Excellent problem-solving skills and a creative mindset to address complex AI challenges. Demonstrated ability to drive research projects from idea to implementation, producing tangible outcomes.
- Published research papers or contributions to the LLM community would be a significant plus.
- Experience with inference tuning and Inference acceleration. Have a deep understanding of GPU and/or other AI accelerators, experience with large scale AI networks, pytorch 2.0 and similar technologies.
- Experience with evaluation of AI systems, LLM application & agent development is desirable.
- Strong understanding of cutting-edge LLM research (e.g., long context, multi modality, alignment research, agent ecosystem, etc.) and possess practical expertise in effectively implementing these advanced systems as a plus
- Strong understanding of distributed computing framework & performance tuning and verification for training/finetuning/inference; Being familiar with PEFT, RL, MoE, CoT or Langchain is a plus