Log in
Sign up
Back to roles
Posted about 13 hours ago
Internship
Redmond, Washington
Hybrid
AI alignment
frontier evaluations
red teaming
post-training
machine learning
foundation models
large language models
reinforcement learning
supervised fine-tuning
preference-based methods
deception
sandbagging
reward hacking
agentic misalignment
Top applicants score 80%+.
See where you land & how to improve
Posted 12 hours ago. Apply today while the pile is still small.
Research Intern - Self-Improving & Secure AI Agents - Microsoft Research
Microsoft
1 day ago
Research Intern - Frontier Tuning
2 days ago
Research Intern - Model Foundations for Knowledge & Reasoning - Microsoft Research