Thinking in public,
building in private.
Engineering notes, product observations, and practical lessons from the work.
is a feature.
What RAG actually needs to work in production
Retrieval quality, evaluation, and product context matter more than picking the most impressive model.
Read note ↗Choosing boring technology for a system that has to last
Why proven tools are often the most ambitious choice when a team has to own the system after launch.
Modernizing a system without a full rewrite
A practical approach to replacing the parts of a legacy system that are actually causing pain.
Evaluation is the product surface AI teams forget
Why useful AI systems need visible ways to measure, correct, and improve their behavior.
The hidden cost of unclear ownership
How unclear responsibility turns small technical decisions into recurring operational drag.
Small tools can carry surprisingly large leverage
Choosing a narrow surface that removes a repeated decision instead of building a broad platform too early.
Designing systems that explain themselves
The operational details that make software easier to trust when something goes wrong.
Building something
difficult?
Tell us what you're building, improving, or trying to figure out.