Resources
Resources for building production feature infrastructure.
Everything on this site, organized by the question that brings people here.
I want to understand the fundamentals
Category education, from definitions to the temporal model.
guide
What is a feature store?
The category from first principles — storage, computation, serving, and where execution fits.
guide
Online vs offline feature stores
The two workloads behind every feature, and what must stay identical between them.
guide
Point-in-time correctness
Event time, availability time, observation time — and how leakage silently inflates backtests.
I have a production ML problem
The specific failure modes ClearFeature is built against.
I am evaluating architecture
The decisions that shape a feature platform adoption.
decision
Build vs buy
What actually goes into a production feature platform — and when building in-house makes sense.
product
The ClearFeature platform
Core, Feature Projects, the DAG, and one execution model across historical and live paths.
deployment
Self-hosted deployment
Running ClearFeature on infrastructure you control — components, boundaries, responsibilities.
I am comparing alternatives
Researched comparisons that say when the alternative is the better choice.
I want to run it
The technical path starts on GitHub, not on this website.