Research Engineer, AI Safety & Alignment

About the role

As a Research Engineer on the Safety team, you will tackle one of the most critical challenges in AI: ensuring models behave in line with human values. You will bridge theoretical research and product implementation, writing production‑grade code to test hypotheses and integrate safety solutions into our consumer AI products.

Key responsibilities

• Develop novel evaluation methodologies and metrics to assess safety and alignment of large language models.
• Research and implement cutting‑edge techniques for model alignment, value learning, and interpretability.
• Conduct adversarial testing to uncover vulnerabilities and failure modes.
• Analyze and mitigate bias, toxicity, and harmful behaviors using RLHF and fine‑tuning.
• Collaborate with engineering and product teams to translate safety research into scalable, production‑ready solutions.

Required profile

• PhD or equivalent experience in Computer Science, Machine Learning, or a related field.
• Strong passion for responsible AI development and solving complex safety challenges.
• Experience writing clear, production‑facing code and working with GPUs for training and serving.
• Familiarity with data pipelines, distributed training, and modern ML techniques such as transformers and reinforcement learning.

Required skills

• Python programming.
• GPU training/serving/debugging.
• Data pipelines and data infrastructure.
• Transformers and reinforcement learning (including RLHF).
• Distributed model training.
• Kubernetes, Docker, and cloud deployment.
• Explainable AI (XAI) and interpretability techniques.