🛠 Tech Stack

💼 About This Role

You'll join the Post-Training team to fine-tune AI models, optimizing their performance and quality. Your work directly contributes to groundbreaking advancements in AI, shaping an era where technology is a companion in daily life. You'll improve model performance and training methods across data, compute, and algorithms.

🎯 What You'll Do

• Develop alignment algorithms and loss functions for data efficiency
• Write data pipelines to process diverse web data
• Identify quality signals to understand model performance
• Design sampling algorithms to improve serving efficiency

📋 Requirements

• at least PhD (or equivalent)
• Experience with GPUs (training, serving, debugging)
• Experience with data pipelines and data infrastructure
• Strong understanding of modern ML techniques (RL, transformers)

✨ Nice to Have

• Experience with product experimentation and A/B testing
• Experience training large models in a distributed setting
• Familiarity with ML deployment and orchestration (Kubernetes, Docker)