菜单
Design an LLM A/B test — pick platform (Statsig or GrowthBook), primary metric, guardrails, sample size with LLM-noise buffer, CUPED, sequential stopping, and multiple-comparison correction. Use when you need help with ab plan.
Choose a 3D generation or reconstruction pipeline given input type, output format, and use case. Use when you need help with 3d pipeline.
|. Use when you need help with 8 bit orbit video template.
Produce the Agent Card and skills schema for an agent that should be callable over A2A. Use when you need help with a2a agent spec.
Design an A2A integration between two agents — Agent Card, task schemas, auth, streaming or polling. Use when you need help with a2a integrator.
Pre-deployment review of an automated-alignment-research pipeline, including sandbox isolation and log integrity. Use when you need help with aar deployment review.
accessibility-designer. Use when you need help with accessibility designer.
| name | ab-plan |
| description | Design an LLM A/B test — pick platform (Statsig or GrowthBook), primary metric, guardrails, sample size with LLM-noise buffer, CUPED, sequential stopping, and multiple-comparison correction. Use when you need help with ab plan. |
| license | CC-BY-NC-SA-4.0 |
| phase | 17 |
| lesson | 21 |
| metadata | {"version":"1.0.0","tags":["ab-testing","statsig","growthbook","cuped","sequential","benjamini-hochberg","srm"]} |
Given the feature change (prompt / model / generation parameter), baseline metrics, expected lift, and team posture (warehouse-native OSS vs bundled SaaS), produce an A/B plan.
Produce:
Hard rejects:
Refusal rules:
Output: a one-page plan with platform, primary + guardrails, sample size, design, CUPED, correction, SRM policy. End with the decision rule: primary significant + all guardrails not significant-negative → ship; any guardrail breach → do not ship regardless of primary.
