Curated roles
Browse roles focused on reducing risks from advanced AI systems, including alignment research, safety engineering, evaluations and related operations.
363 active roles found.
Associate role on Faculty’s AI Safety team supporting delivery of frontier model evaluations, AI safety red teaming, and related client projects for government and industry.
Internship focused on ML research engineering for LLM and agent evaluation, red-teaming, guardrails, and model safety/robustness tooling.
Research Scientist focused on interpretability and techniques for understanding and steering large AI models, with explicit connection to making models safer and supporting safe model development.
Operations and program lead for Harvard-MIT AI Safety Workshops and related AI safety community events, supporting AI safety career development and workshop programming rather than doing research directly.
Research Scientist role focused on AI safety research, model behavior studies, interpretability, and building reproducible tooling for safe AI deployment.
Assistant AI Security Software Engineer focused on AI security research, AI red teaming, adversarial machine learning, and building tools for AI security applications.
Research role on OpenAI’s Chat and Multimodal Safety team focused on multimodal safety research, safety evaluations, post-training, and interventions to ensure frontier models behave safely.
Subject matter expert role focused on child safety enforcement for generative AI models, including evaluations, red-teaming, detection systems, and release gating for harmful content.
Staff software engineer building web applications and backend systems for AI model evaluation, real-time threat detection, and adversarial red teaming for frontier AI deployments.
Senior full-stack software engineer role building AI model evaluation, threat detection, and adversarial red teaming products for frontier labs.
Machine Learning Engineer focused on AI safety solutions, including adversarial testing, model evaluation, robust inference, and monitoring for secure deployment of AI systems.
Research engineer role focused on AI safety evaluations, interpreting model behavior for misalignment, and implementing controls for autonomous AI deployments.
Lead multidisciplinary research teams at the AI Security Institute to advance frontier AI safety and risk mitigation, including evaluations, benchmarks, and safeguards for advanced AI systems.
Software engineering role on Perplexity’s Model Behavior team focused on prompt/context engineering, model behavior shaping, failure-mode analysis, and some evaluation work for LLM systems.
Lead role overseeing GenAI safety research and delivery, including adversarial testing, risk evaluations, red teaming methodology, and client-facing safety deliverables.
Research scientist role on DeepMind’s Safety Oversight team focused on monitoring deployed AI models for safety and alignment, detecting misuse and misbehavior, and building automated evaluation and monitoring systems.
Senior virology role focused on biosecurity mitigation, safety frameworks, and advanced evaluations for frontier AI models to prevent biological misuse.
Research Scientist role focused on AI evaluation, language model understanding, robustness, red teaming, and alignment for LLM evaluation infrastructure.
Senior research manager for an AI safety organization, leading empirical deep learning research on LLMs and multimodal models, managing researchers, and contributing to safety-focused papers.
Infrastructure engineer for Anthropic’s Safeguards research team, building tooling, pipelines, and evaluation/scoring workflows that support misuse detection and safety research for frontier AI systems.