Reference ledger · curriculum v1.0
AI Engineering Glossary
Precise working definitions for the systems you build. Start with the meaning, then use the examples, distinctions, and lesson links to turn vocabulary into judgment.
Each entry separates a term's working definition from why it matters in an AI system, how it appears in practice, and what engineers commonly confuse it with. Search by term, alias, or idea; narrow the ledger by learning area; then follow the linked course lesson or primary source when you need the full derivation. Every entry has a stable fragment link, so you can cite one definition directly without sending someone through the entire index.
Reference entries
Loading reference entries...
All 250 terms by learning area
237 terms have their own page with the definition, why it matters, common confusion and course links.
Math & training
- Activation Checkpointing
- Activation Function
- Adam (Optimizer)
- AdamW
- Autograd
- Backpropagation
- Batch Size
- Contrastive Learning
- Cross-Entropy
- Data Augmentation
- DPO (Direct Preference Optimization)
- Dropout
- Eigenvalue
- Epoch
- Fine-tuning
- Gradient
- Gradient Accumulation
- Gradient Clipping
- Gradient Descent
- Hyperparameter
- JAX
- Knowledge Distillation
- Learning Rate
- Learning Rate Schedule
- LoRA (Low-Rank Adaptation)
- Loss Function
- Mixed Precision
- NaN (Not a Number)
- Normalization
- Optimizer
- Overfitting
- QLoRA
- ReLU
- RLHF (Reinforcement Learning from Human Feedback)
- SFT (Supervised Fine-Tuning)
- Softmax
- Stochastic Gradient Descent (SGD)
- Transfer Learning
- Underfitting
- Warmup
- Weight
- Weight Decay
Models & inference
- Attention
- Autoregressive
- CNN (Convolutional Neural Network)
- CUDA
- Decoder
- Decoding Strategy
- Diffusion Model
- Encoder
- GAN (Generative Adversarial Network)
- GPT
- Inductive Bias
- Inference
- KV Cache
- LLM (Large Language Model)
- Logits
- MoE (Mixture of Experts)
- Nucleus Sampling (Top-p)
- Parameter
- Perplexity
- Quantization
- Self-Attention
- Speculative Decoding
- Stop Sequence
- Streaming
- Temperature
- Time to First Token (TTFT)
- Top-k Sampling
- Transformer
- VAE (Variational Autoencoder)
Data & representations
Retrieval & generation
Prompting & context
Agents & tools
- Agent
- Agent Harness
- Agent Memory
- Agent Skill
- Agent State
- Approval Gate
- Checkpoint
- Compensating Action
- Delegation
- Durable Execution
- Function Calling
- Human-in-the-Loop (HITL)
- MCP (Model Context Protocol)
- Multi Round-Trip Request (MRTR)
- Orchestration
- Planning
- ReAct
- Sandbox
- Skill Bundle
- Skill Catalog
- Skill Discovery
- Skill Invocation
- Stateless MCP
- Structured Output
- Swarm
- Termination Condition
- Tool Contract
Evaluation & safety
AI-native development
- Backpressure
- Circuit Breaker
- Coding Agent
- Cost per Successful Task
- Flaky Test
- Handoff
- Idempotency
- Model Router
- Observability
- Patch
- Progressive Disclosure
- Rate Limit
- Regression Test
- Repository Instructions
- Repository Map
- Reproducible Build
- Retry with Backoff
- Reviewer Agent
- Scope Contract
- Semantic Cache
- Test Oracle
- Trace
- Worktree
Infrastructure & serving
Reliability & operations
Security & governance
- AI Risk Assessment
- Audit Log
- Content Provenance
- Data Classification
- Data Exfiltration
- Data Lineage
- Data Minimization
- Datasheet for Datasets
- Defense in Depth
- Indirect Prompt Injection
- Jailbreak
- Membership Inference
- Provenance Attestation
- Purpose Limitation
- Red Teaming
- Separation of Duties
- Software Bill of Materials (SBOM)
- Threat Model
- Trust Boundary
- Zero Trust