Causal Methods
Time-series causal inference and statistical validation methods.
- Applied Causal Inference in progress
- double_ml_time_series in progress
- TemporalValidation / temporalcv released
Brandon Behring
Building applied causal methods, AI evaluation tooling, and a study notes corpus.
Live now
What I'm building now
AI Evaluation
Methodology-led AI safety evaluation: real evals, calibration, anti-overclaim discipline.
- prompt-injection detection: OOD generalization study released
- eval-toolkit in progress
Course Notes
Long-form study notes synthesized from technical courses. DLAI is the first; more course-derived notes lined up.
- DLAI Study Notes in progress
About
Currently working through post-transformer sequence model architectures (SSMs, Mamba, Hyena, DeltaNet) and reinforcement learning + optimal control. Expanding the course-derived study notes pattern beyond DLAI. Direction: research engineering on next-generation AI architectures, with applied causal methods, risk analysis, and rigorous evaluation as the methodology spine.
Methodology, AI disclosure, and audit trail: How this was made →
Earlier peer-reviewed research: Publications →
Contact
- Email: brandon.m.behring@gmail.com
- GitHub: @brandon-behring
- LinkedIn: /in/brandon-behring