Large Language Model Training System Engineer Graduate (Applied Machine Learning) - 2027 Start

What you'll do

• Develop training systems for large-model post-training, including SFT and RL.
• Design elastic, multi-tenant training across data centers and heterogeneous hardware, improving throughput and stability.
• Design reinforcement learning systems and user-friendly training interfaces to improve training efficiency.

What they ask for

• Completing or recently completed a Bachelor's or Master's degree in Computer Science or a related discipline.
• Proficient in one or more programming languages such as Python, Rust, or C++; able to write clean code and demonstrate framework design and abstraction skills.
• Experience in training framework design or training system optimization.
• Strong interest in investigating technical problems, low-level principles, and performance bottlenecks, and applying scientific modeling methods.

Nice to have

• Participation in large-model training engineering or complex distributed-system development; experience in computer architecture optimization, heterogeneous hardware, or high-performance networking.
• Scientific optimization skills, including analyzing training efficiency; familiarity with mainstream reinforcement learning frameworks and training frameworks.

Tools

Python, Rust, C++, SFT, RL

AI focus

Builds large-language-model post-training systems, including supervised fine-tuning and reinforcement learning, and improves training efficiency.