TikTok is the leading destination for short-form mobile video, and they are seeking Software Engineering Interns to join the Model Infra team. The role focuses on optimizing the performance of recommendation systems and involves engaging in impactful short-term projects that provide real-world experience.
Responsibilities
Drive the optimization of training and inference pipelines to maximize hardware utilization (MFU/HFU) for models featuring hundreds of billions of dense parameters
Architect specialized systems to support the integration of LLMs into the recommendation stack, focusing on memory-efficient attention mechanisms and advanced KV cache management for long-sequence user modeling
Build and optimize high-concurrency engines for Petabyte-scale streaming training, handling continuous parameter updates and high-frequency data ingestion without compromising stability
Work closely with researchers to design next-generation recommendation architectures optimized for modern GPU/NPU interconnects, ensuring high-bandwidth utilization across the cluster
Innovate on how we store and synchronize massive model states across heterogeneous memory hierarchies (HBM, DDR, and NVMe)
Qualification
Required
Currently pursuing an Undergraduate/Master in Software Development, Computer Science, Computer Engineering, or a related technical discipline
Strong programming skills in C++ and Python
Solid understanding of Computer Architecture and the GPU software stack (CUDA, Triton, or NCCL)
Experience with deep learning frameworks (e.g., PyTorch, TensorFlow) and a desire to 'look under the hood' of model execution runtimes
A strong interest in solving system-level bottlenecks in large-scale distributed environments
Preferred
Experience with Transformer-based architectures, 3D parallelism (TP/PP/DP)
Deep understanding of the torch.compile stack, including TorchDynamo (graph acquisition) and TorchInductor (lowering)
Hands-on experience writing high-performance kernels or optimizing collective communication (e.g., customizing NCCL/UCX)
Familiarity with RDMA networking, high-performance storage, or specialized Parameter Server architectures
Success in programming competitions (ACM-ICPC) or contributions to prominent open-source AI infrastructure or high-performance computing projects
Benefits
Interns have day one access to health insurance, life insurance, wellbeing benefits and more.
Interns also receive 10 paid holidays per year and paid sick time (56 hours if hired in first half of year, 40 if hired in second half of year).
Interns who are not working 100% remote may also be eligible for housing allowance.
TikTok is a short-form video entertainment app and social network platform. It is a sub-organization of ByteDance.
