ai-evaluation
Skill1 starsChanged 6 months agoMITView on GitHub
Practical knowledge for evaluating AI/LLM systems. Covers language modeling metrics (perplexity, cross-entropy), exact evaluation methods (functional correctness, similarity), AI-as-judge patterns, comparative evaluation, evaluation criteria for production systems, model selection workflows, and designing end-to-end evaluation pipelines.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- SWEStash/swe-workflow-skillstheir own version, first seen Nov 8, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- SWEStash/swe-workflow-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository21 other files it gives its agents.
- ai-context-engineeringSkill
- ai-context-engine-productionSkill
- ai-context-engineSkill
- ai-dataset-engineeringSkill
- ai-finetuningSkill
- ai-foundation-modelsSkill
- ai-inference-optimizationSkill
- ai-multi-agent-mcpSkill
- ai-production-architectureSkill
- ai-prompt-engineeringSkill
- ai-rag-and-agentsSkill
- ai-rag-defenseSkill
- ddia-architectureSkill
- ddia-batch-stream-processingSkill
- ddia-data-ethicsSkill
- ddia-data-modelingSkill
- ddia-replication-shardingSkill
- ddia-transactions-consistencySkill
- job-description-analyzerSkill
- resume-content-writerSkill
- resume-fit-scorerSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 21
Check guidelines.md to find which files to load for your task
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AI Evaluation
Knowledge from "AI Engineering" by Chip Huyen (Chapters 3-4). Practical methods for evaluating foundation models and AI systems built on top of them.
Quick Start
- Check
guidelines.mdto find which files to load for your task - Load only relevant files (each topic has knowledge.md, rules.md, examples.md)
- Apply guidance to your work
Contents
References
| Category | Purpose |
language-modeling-metrics | Entropy, cross-entropy, perplexity, bits-per-character |
exact-evaluation | Functional correctness, exact match, lexical/semantic similarity, embeddings |
ai-as-judge | When to use AI judges, how to prompt them, limitations and biases |
comparative-evaluation | Ranking models with pairwise comparisons, Bradley-Terry, scalability challenges |
evaluation-criteria | Domain capability, generation (factual, safety), instruction-following, cost/latency |
model-selection | Selection workflow, open source vs API, navigating public benchmarks |
evaluation-pipeline | End-to-end pipeline design, scoring rubrics, evaluation methods |
Workflows
| Task | Workflow |
| Choose a model (build vs buy, OS vs API) | workflows/select-model.md |
| Design an end-to-end evaluation pipeline | workflows/design-eval-pipeline.md |
Guidelines
See guidelines.md for task-based file selection.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.