Codenames Clue Generator v2: Wikipedia Graphs, Gemini Embeddings, and What Still Doesn’t Work
Building the graph-based approach I promised in v1 — and discovering the yield gap isn’t where I expected it to be.
Traffic Routing
Optimizing Transaction Routing with Linear Programming
ML Model Monitoring Metrics
Detecting silent model failures early and consistently
Distribution Drift Tolerance
Identifying your personal preference for measuring population drift
PyMC Wrapper
Making PyMC models more accessible and reusable
MMM: Miracle of Marketing Measurement or Misleading Modeling Methodology?
MMMs are a future-proofed tool for measuring marketing effectiveness in a world of increased online privacy, but can be prone to multiple silent failure methods that provide inaccurate results.
Spaghetti Code
Making generative noodle art
Codenames Clue Generator using Semantic Similarity
Utilizing Tensorflow pre-trained embeddings to recommend potential clues to the codemasters in the card game Codenames
Multi-Armed Bandits Exploration
Benchmark Comparisons and UCB Visualization
Basic Geometric Tiling
Inspiration from Italian Mosaics