Five levels, one price each
Free, $19, $49, $99 and $149, each paid once in our store. Every level includes the ones before it, so you can stop where the job is done. L4 SCALE is the complete series.
Roadmap · S2 · AI Evaluation & Reliability
Turn 'it looked fine in the demo' into evidence you can compare.
You get criteria the team can defend: acceptable variation defined, runs compared like for like, uncertainty written down, and an evaluation that CI and owners can act on.
Who it is for: Engineers, QA leads and AI owners who have to decide whether an LLM or RAG feature behaves well enough to ship.
Buying with confidence
What you paid for any lower level of the same series counts toward any higher level (L1 START at $19 makes L4 SCALE cost $130). How to claim it
Paid levels carry a 7-day money-back guarantee and at least six updates a year for five years, as set out in the terms.
Follow the five levels in order
Tick a level when you finish it. Progress is stored in this browser only; no account or email is needed.
L0LLM Failure Pattern ScanL0 FREE · Free
Stop trusting one good answer: repeat a small trial and see which of five failure patterns show up.
Best for: Engineers about to evaluate an LLM feature who are not sure where to start.
Price: Free, no payment.
Get it free in our store Also on Gumroad Open LLM Failure Pattern Scan product page
L1Initial LLM Test ChecklistL1 START · $19
Run five repeatable checks on a promising LLM demo and write down the variation and the risks.
Best for: QA engineers and technical leads starting LLM testing who want a short, repeatable routine.
Price: $19, paid once. Includes L0-L1.
7-day money-back guarantee. How the upgrade credit works
Get L1 START for $19 in our store Also on Gumroad Open Initial LLM Test Checklist product page
L2Non-Deterministic Evaluation DesignL2 UNDERSTAND · $49
Define how much variation is acceptable and compare runs on evidence the team can defend.
Best for: Engineers who own AI quality criteria and regression decisions and are tired of subjective arguments.
Price: $49, paid once. Includes L0-L2.
7-day money-back guarantee. How the upgrade credit works
Get L2 UNDERSTAND for $49 in our store Open Non-Deterministic Evaluation Design product page
L3Production AI Evaluation SystemL3 APPLY · $99
Move evaluation out of the notebook: dataset, evaluator, thresholds, CI and monitoring that owners can act on.
Best for: Teams that run AI evaluations beyond a one-off experiment and need a signal that survives the demo.
Price: $99, paid once. Includes L0-L3.
7-day money-back guarantee. How the upgrade credit works
Get L3 APPLY for $99 in our store Open Production AI Evaluation System product page
L4Enterprise Evaluation Platform BlueprintL4 SCALE · $149
Set up shared evaluation governance across teams, with a separate risk level and threshold for each use case.
Best for: Platform and quality leaders who standardize AI evaluation across teams and need local differences to stay visible.
Price: $149, paid once. Includes L0-L4, the complete series.
Why L4: L4 SCALE is $50 more than L3 and adds the blueprint for shared evaluation governance across teams; every level below it is included.
7-day money-back guarantee. How the upgrade credit works
Get L4 SCALE for $149 in our store Open Enterprise Evaluation Platform Blueprint product page
Compare before you choose
The workflows page shows all 25 workflows with sample reports, videos and the terms. Compare the levels of S2 · AI Evaluation & Reliability, or see the other roadmaps.