Sturdy Statistics Blog – Sturdy Statistics

The Performance of Clarity: Why We Chose C for Our Statistical Engine

At Sturdy Statistics, our models are probabilistic, structured, and deeply recursive — Bayesian DAGs that encode how information flows through the data. Our models are not…

The Quiet Power of SQL

Early in our company’s life, we built everything around modern data frameworks — until we realized the simplest, most reliable tool had been in front of us all along.

Bottlenecks, Pure Functions & a Newborn: Why We Migrated from Python to Clojure

At Sturdy Statistics, our mission has always been to make sense of complex data — whether it comes from genomes or text. Our core technology began as something unusual: a gen…

Bayesian Hierarchical Models: Structured Uncertainty for Hierarchical Data

In Part I, we saw how hypernetworks let neural nets adapt to new datasets by learning embeddings and generating dataset-specific weights. That approach worked surprisingly…

Hypernetworks: Neural Networks for Hierarchical Data

Neural nets assume the world is flat. Hierarchical data reminds us that it isn’t.

Cutting DuckDB Storage in Half with SciPy-Style Sparse Arrays

I was surprised to discover that nearly three-quarters of our storage was near-zero float32s — and motivated to take advantage of this sparsity.

AI Keeps Shifting Right: Coping with the Limitations of Large Language Models

Software reliability has improved dramatically thanks to a popular trend called shifting left. By pushing testing, security checks, and quality assurance earlier in the…

Product Reviews Don’t Line Up With Ratings

It’s common knowledge but the juxtaposition is shocking.

How Sturdy Statistics Works: An Alternative Foundation for AI

Sturdy Statistics builds interpretable AI models that understand structure, quantify uncertainty, and work at any scale.

Michael McCourt