Member of Technical Staff, Safety for Agents
Core
Developing safety mechanisms, data generation, and evaluation methods for Large Language Models (LLMs) that can access external resources and take actions.
Role type
Senior IC machine learning engineer (safety for agents)
Builds
Next generation of safe, trustworthy, and secure LLMs capable of autonomous actions
Domain
AI safety, large language models, distributed training
Deliverable
production ML models
Required skills
statistical analysis, software engineering, experimental design, data collection management, LLM training on distributed infrastructure, ML system robustness evaluation, Python programming, PyTorch/TensorFlow/JAX frameworks, academic research publication
Preferred skills
experience with human annotators, bias analysis in datasets, generalizability improvement
Technologies
Python, PyTorch, TensorFlow, JAX
Responsibilities
Design and conduct data collection tasks with human annotators, implement post-training algorithms for safety, evaluate model performance and dataset quality, collaborate with cross-functional ML and data teams
Seniority
Senior, hands-on IC
