RLHF Specialist

About the role

An RLHF Specialist improves and aligns AI language models using Reinforcement Learning from Human Feedback, designing pipelines that boost performance, safety, factual accuracy, and alignment with human values.

Key responsibilities

• Generate high-quality preference data by comparing multiple model responses and ranking them based on helpfulness, honesty, and harmlessness.
• Design complex multi‑turn prompts to stress‑test model behavior and expose reasoning or safety weaknesses.
• Write detailed chain‑of‑thought explanations to train reward models on why specific responses are superior.
• Collaborate with ML engineers to analyze failure modes, identify data gaps, and improve reinforcement learning outcomes.
• Develop and iterate annotation strategies for preference scoring, ensuring consistency across a global team.

Required profile

• Minimum 2 years experience in data annotation, model evaluation, computational linguistics, or trust and safety.
• Strong proficiency in Python and deep‑learning frameworks such as PyTorch, JAX, or TensorFlow.
• Deep understanding of reinforcement‑learning concepts (PPO, trust regions, reward hacking) applied to language generation.
• Hands‑on experience fine‑tuning open‑source models (Llama 2/3, Mistral, Gemma) using LoRA/QLoRA.

Required skills

• Python
• PyTorch
• TensorFlow
• JAX
• LoRA/QLoRA fine‑tuning techniques
• Annotation tools (LabelBox, Scale AI, Snorkel)
• AWS SageMaker
• GCP Vertex AI