AdTech AdTalk Ask an Expert (ft. Tylynn Pettrey, Chalice AI)
Jan 30, 2026
Tylynn Pettrey, SVP of Data Science at Chalice AI, leads work on predictive modeling, deep learning, and low-latency ML for advertising. She traces modeling history from linear regression to transformers. Short takes cover sparse ad data, embeddings, model explainability, agentic AI hype, quantum in ad tech, practical hiring skills, and a quirky mass spectrometer pandemic story.
AI Snips
Chapters
Books
Transcript
Episode notes
Linear Regression: The Origin Of Predictive Lines
- Linear regression began in the 1800s as a way to fit a best line to noisy observations and predict continuations.
- Tylynn Pettrey emphasizes minimizing collective error (mean squared error) to choose the best predictive line.
Logistic Regression Maps Probabilities To An S‑Curve
- Logistic regression maps continuous inputs to an S-curve to produce class probabilities for yes/no outcomes.
- Tylynn explains it's fast and useful for millisecond classification like click prediction.
Decision Trees As Flowchart Reasoning
- Decision trees model expert-like if/then reasoning as flowcharts and can be ensembled for stronger predictions.
- Ensembling aggregates many decision paths to fill missing information from historical patterns.
