Software Engineering
▼
AI Safety Researcher / Alignment Engineer
Master alignment theory, use AI agents to accelerate research velocity.
01Prerequisites
Linear Algebra Foundations
+
Vectors, matrices, eigenvalues for ML fundamentals.
Probability and Statistics
+
Distributions, Bayes theorem, hypothesis testing essentials.
Python Programming
+
Fluent Python for research, scripting, experimentation.
02Core Concepts
Deep Learning Fundamentals
Neural networks, backprop, transformers for alignment study.
Alignment and Safety Theory
Specification gaming, reward hacking, corrigibility concepts.
Mechanistic Interpretability
Understanding neural network internals, circuits, attention.
Evaluating and Testing AI Systems
Red teaming, adversarial testing, benchmarking rigor.
tap any prerequisite node to explore →