AI Cheat Sheets

One-page AI cheat sheets for developers: how each mechanism works, the rules that hold, and when to use it. Click a sheet to watch it build step by step.

  • Home /
  • AI Cheat Sheets

Benchmark Contamination cheat sheet

A score proves capability only while the test stayed unseen; once questions leak into training, it measures memory.

#Model EvaluationAnimated

→ Topic: Benchmark Contamination

Benchmark Datasets cheat sheet

A public score only narrows the shortlist; a private test built from your own work makes the call.

#Synthetic Data & GenerationAnimated

→ Topic: Benchmark Datasets

Bias and Fairness Metrics cheat sheet

When groups have different base rates, you choose which fairness to satisfy and write down the trade-off.

#Safety & Red TeamingAnimated

→ Topic: Bias and Fairness Metrics

Browser and Computer Use Agents cheat sheet

Driving the screen reaches software that has no API, but actions that cannot be undone still need a human to confirm them.

#AI Agent ArchitectureAnimated

→ Topic: Browser and Computer Use Agents

Catastrophic forgetting in fine-tuning cheat sheet

Catastrophic forgetting is the loss of earlier skills after further training. Compare general and task performance before and after fine-tuning.

#LLM Training & Pre-Training

→ Read the article

Class Imbalance cheat sheet

A model trained on skewed data can look accurate while missing the rare cases; change the metric before you change the data.

#Safety & Red TeamingAnimated

→ Topic: Class Imbalance

Code Execution Agents cheat sheet

The model writes the program, the sandbox contains it; the result returns as the next observation.

#AI Agent ArchitectureAnimated

→ Topic: Code Execution Agents

Code LLMs cheat sheet

A code LLM predicts likely code from patterns, so read and run every draft before you trust it.

#AI Coding AssistantsAnimated

→ Topic: Code LLMs

Confusion Matrix cheat sheet

Every precision, recall or accuracy figure is arithmetic on four cells; read them per class before trusting the total.

#Model EvaluationAnimated

→ Topic: Confusion Matrix

Constitutional AI Prompting cheat sheet

The loop checks a draft against the stated rules; it never checks whether the answer is true.

#Prompt EngineeringAnimated

→ Topic: Constitutional AI Prompting

Context Engineering for Code cheat sheet

The assistant has no memory of your project; whatever its context window doesn't show it, it fills in with generic defaults.

#Agentic & Autonomous CodingAnimated

→ Topic: Context Engineering for Code

Context engineering for coding agents cheat sheet

Context engineering for coding agents means choosing what an AI coding tool reads before it writes, so its code fits your project. A plain-language guide.

#Agentic & Autonomous Coding

→ Read the article

Context Window Management cheat sheet

A bigger window raises the ceiling, but each call still has to choose which tokens deserve a slot.

#LLMOps & ProductionAnimated

→ Topic: Context Window Management

Contextual Retrieval cheat sheet

Contextual retrieval pays once at indexing time so every chunk carries the meaning it lost when the document was split.

#Retrieval-Augmented GenerationAnimated

→ Topic: Contextual Retrieval

Continuous Batching cheat sheet

Finished requests free their slot at once, so the GPU stops waiting for the slowest request in the batch.

#Inference OptimizationAnimated

→ Topic: Continuous Batching

Convolutional Neural Network cheat sheet

Learned filters scan small patches, pooling keeps the strongest signals, and stacking layers turns edges into whole objects.

#Neural Network ArchitecturesAnimated

→ Topic: Convolutional Neural Network

Data Augmentation cheat sheet

Augmentation adds variety to the data you already have, but only when every transform keeps the label true.

#Training Data Quality & CurationAnimated

→ Topic: Data Augmentation

Data Deduplication cheat sheet

Each filter catches copies the previous one missed, and a threshold set too loose deletes rare examples along with real duplicates.

#Training Data Quality & CurationAnimated

→ Topic: Data Deduplication

Data Drift cheat sheet

A drift alert says the data moved, not that accuracy fell, so verify before you retrain.

#Safety & Red TeamingAnimated

→ Topic: Data Drift

Data Labeling and Annotation cheat sheet

The model learns whatever the answer key says, so fix the guidelines before you label more.

#Training Data Quality & CurationAnimated

→ Topic: Data Labeling and Annotation

Data Leakage cheat sheet

A transformation fit before the split grades the model on answers it has already seen.

#Safety & Red TeamingAnimated

→ Topic: Data Leakage

Data Poisoning cheat sheet

Training-time attacks need retraining to fix; RAG and agent memory stay open after deployment.

#Safety & Red TeamingAnimated

→ Topic: Data Poisoning

Data Preprocessing cheat sheet

Fit every transform on the training rows only and keep the steps as one pipeline that also runs on new data.

#Training Data Quality & CurationAnimated

→ Topic: Data Preprocessing

Data Versioning cheat sheet

A pointer in Git plus bytes in cheap storage rebuild any past dataset, but the hash cannot tell good data from bad.

#Safety & Red TeamingAnimated

→ Topic: Data Versioning

Dataset Bias cheat sheet

Collecting more data from the same source only enlarges the skew; a representative sample is what fixes it.

#Safety & Red TeamingAnimated

→ Topic: Dataset Bias

Decoder-Only Architecture cheat sheet

The model reads prompt and reply as one stream and adds one token per pass, so the causal mask decides what each token may see.

#Transformer & Attention InternalsAnimated

→ Topic: Decoder-Only Architecture

Diffusion Models cheat sheet

Quality comes from many small denoising steps, so slowness and fine-detail errors are built into the design.

#AI Image Generation & EditingAnimated

→ Topic: Diffusion Models

Document Parsing and Extraction cheat sheet

OCR only reads the characters; parsing has to rebuild headings, tables and reading order, and when it fails, nothing raises an error.

#Knowledge Retrieval SystemsAnimated

→ Topic: Document Parsing and Extraction

Domain-Specific Prompting cheat sheet

Vocabulary, role and constraints steer a general model toward a field, but they change its behavior, not its competence.

#Prompt EngineeringAnimated

→ Topic: Domain-Specific Prompting

ELO Rating for LLMs cheat sheet

An arena rank shows which answers people preferred; inside the confidence interval, a rank change is a tie.

#Model EvaluationAnimated

→ Topic: ELO Rating for LLMs