Research Scientist Graduate - (Trust and Safety, CV/NLP/Multimodal LLM) - 2026 Start(PhD)

Written by the T&S Network from TikTok's posting.

Builds computer vision, NLP and multimodal models that detect content and behavior violating TikTok's community guidelines and related regulations. The algorithm team works on recognizing violating content across the platform and on the distributed training framework that supports these large models.

What you'd do

• Develop computer vision or multimodal models that recognize violating content on TikTok.
• Explore large multimodal and vision models such as CLIP, BLIP, Flamingo and ViT-22B for violation detection.
• Explore LLM applications in business scenarios, including pre-training, zero-shot and few-shot learning, and hard case mining.
• Continuously optimize the distributed training framework to suit large model training.

What it asks for

• Has finished or is close to finishing a PhD in computing, engineering or a closely related technical field.
• Research experience in at least one of computer vision, multimodality or LLMs.
• Proficiency with at least one deep learning framework such as PyTorch or TensorFlow.
• Strong analytical, problem-solving, communication and collaboration skills.
• Publications in accredited venues such as CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, TPAMI or IJCV (preferred).

Worth knowing

• Must commit to an onboarding date by end of 2026.

Our best summary of the posting, and it can be wrong. The employer's posting is what counts: read it in full on TikToknorth_east before you apply. If we got something wrong, report this listing.