Multimedia AI Intern – LLM Post-training

Job Description

LLM Post-training Engineer Intern (Research & Product) | TikTok

The Tone:
This is an internship at TikTok, a leading global platform for short-form mobile video. The Multimedia AI team focuses on building, researching, and applying Large Language Models (LLMs) to power TikTok’s products. This role is crucial because post-training is where AI research directly impacts real-world deployment, allowing interns to contribute to transforming experimental AI into production-ready applications like video understanding and content classification. Interns will develop fundamental skills and gain hands-on experience by applying their knowledge in practical scenarios.

The TL;DR
• Role: Internship
• Type: Temporary
• Location: In-person, US
• Pay: $45 hourly
• Team: Multimedia AI team
• Mission: Support the development and optimization of LLM post-training strategies to improve model performance, helpfulness, and safety for global products.
• Tech Stack: Python, PyTorch, JAX, DeepSpeed, Megatron-LM

What You’ll Actually Do
• Strategy Development: Support the development and optimization of post-training strategies, including instruction tuning, preference tuning (SFT/DPO/PPO), and model alignment.
• Evaluation Pipeline: Assist in building robust evaluation pipelines to measure model performance, helpfulness, and safety across diverse multimedia product use cases.
• Research Implementation: Participate in the research and implementation of cutting-edge methodologies in reward modeling and human preference learning.
• Production Bridge: Collaborate with engineering teams to bridge the gap between experimental research and production-ready AI applications.
• Data Analysis: Analyze and process large-scale datasets to identify patterns that improve model behavior and alignment quality.

The Must-Haves
• Background: Currently pursuing an Undergraduate or Master’s degree in Computer Science, Machine Learning, or a related technical discipline. This is a student-level role.
• Experience: Not specified.
• Skills: Strong programming skills in Python; experience with deep learning frameworks such as PyTorch or JAX; foundational understanding of Transformer architectures and LLM training principles; demonstrated ability to learn quickly.
• Bonus: Previous experience or research projects involving LLM fine-tuning, RLHF, or synthetic data generation; familiarity with distributed training tools (e.g., DeepSpeed, Megatron-LM); experience or interest in multimodal AI (integrating text with video or audio); a proven track record of building and deploying AI projects.