Glossary

Key AI, LLM, and agent terms defined in plain language. Bookmark this page as a quick reference while building production AI systems.

Agent: An AI system that can autonomously plan, use tools, and take actions to accomplish a goal. Agents combine an LLM with tool-calling, memory, and control flow.; See also: Internal AI OS
RAG: Retrieval-Augmented Generation. A technique that retrieves relevant documents or data before generating an answer, grounding LLM outputs in factual sources.; See also: RAG Over Company Docs
Embedding: A numerical vector representation of text that captures semantic meaning. Embeddings enable similarity search and are the foundation of vector databases.; See also: RAG Over Company Docs
Fine-tuning: The process of further training a pre-trained LLM on domain-specific data to improve its performance on a particular task or style.
Prompt Engineering: The practice of designing input prompts to elicit desired outputs from LLMs. Includes techniques like few-shot examples, chain-of-thought, and system instructions.
Chain-of-Thought: A prompting technique that asks the LLM to reason step-by-step before giving a final answer, improving accuracy on complex tasks.
Few-Shot: A prompting strategy that includes a small number of example input-output pairs in the prompt to guide the model's behavior.
Zero-Shot: Asking an LLM to perform a task without providing any examples, relying entirely on its pre-trained knowledge and instruction following.
Tool Use: The ability of an LLM to call external tools, APIs, or functions during generation. The model outputs a structured tool call, and the runtime executes it.
Function Calling: A structured mechanism where the LLM outputs a function name and arguments as JSON, which the application executes and feeds results back to the model.
Context Window: The maximum number of tokens an LLM can process in a single request, including both input and output. Larger windows allow more data but cost more.
Token: The basic unit of text processed by an LLM. A token is roughly 3/4 of a word in English. Models charge per token for both input and output.
Temperature: A parameter controlling randomness in LLM outputs. Lower values (0.0-0.3) produce more deterministic outputs; higher values (0.7-1.0) increase creativity.
Top-P: Also called nucleus sampling. A parameter that limits token selection to the smallest set whose cumulative probability exceeds P, controlling diversity.
Hallucination: When an LLM generates plausible-sounding but factually incorrect or fabricated information. A major challenge addressed by grounding and RAG.; See also: RAG Over Company Docs
Grounding: Techniques that anchor LLM outputs to real, verifiable sources. Includes RAG, citation requirements, and tool-based fact-checking.; See also: RAG Over Company Docs
Vector Database: A database optimized for storing, indexing, and querying high-dimensional vectors (embeddings). Enables fast similarity search for RAG systems.
Semantic Search: Search based on meaning rather than exact keyword matching. Uses vector embeddings to find conceptually similar content.; See also: RAG Over Company Docs
Chunking: The process of splitting documents into smaller segments for embedding and retrieval. Chunk size and overlap affect retrieval quality.; See also: RAG Over Company Docs
Knowledge Graph: A structured representation of entities and their relationships. Can augment RAG systems by providing structured context alongside vector search.
RLHF: Reinforcement Learning from Human Feedback. A training method where human preferences are used to fine-tune an LLM's behavior via a reward model.
DPO: Direct Preference Optimization. A simpler alternative to RLHF that fine-tunes models directly from preference data without a separate reward model.
LoRA: Low-Rank Adaptation. A fine-tuning technique that adds small trainable matrices to a frozen model, drastically reducing the resources needed for customization.
Quantization: Reducing the precision of model weights (e.g., from FP16 to INT4) to shrink model size and memory requirements with minimal quality loss.
GGUF: GPT-Generated Unified Format. A file format for quantized models optimized for CPU inference with llama.cpp. Popular for local deployment.
KV Cache: A memory optimization that stores previously computed key-value attention pairs, avoiding redundant computation during autoregressive generation.
Attention Mechanism: The core operation in transformers that allows each token to attend to all other tokens, learning which parts of the input are relevant to each output.
Transformer: The foundational neural network architecture for modern LLMs. Uses self-attention to process sequences in parallel, enabling scalable training.
Mixture of Experts: A model architecture where different sub-networks (experts) specialize in different inputs, and a gating network routes each input to the most relevant experts.
Agent Loop: The core execution pattern of an AI agent: observe the environment, think about next steps, take an action, and repeat until the task is complete.; See also: Internal AI OS
LLM: Large Language Model. A model trained on large text corpora to predict the next token. LLMs are the foundation of modern AI agents and workflows.
Prompt: The input given to a model to elicit a response. A prompt combines the instruction, context, and any examples the model needs.
System Prompt: Instructions provided at the start of a conversation that set the model's role, tone, and constraints for every turn.
Inference: The process of running a trained model to generate output, as opposed to training. Inference is what happens when an LLM answers a request.
Autoregressive: A generation mode where the model produces one token at a time, each conditioned on all previously generated tokens.
Tokenization: The process of splitting text into tokens that a model can process. A token is roughly three-quarters of a word in English.
MCP: Model Context Protocol. An open standard for connecting LLMs to external tools, resources, and prompts in a structured way.; See also: Internal AI OS
Orchestration: Coordinating multiple LLM calls, tools, or agents to complete a larger task, including decomposition, delegation, and result synthesis.
Multi-Agent System: A system where multiple specialized agents coordinate, often through an orchestrator, to solve tasks that a single agent cannot.
ReAct: Reasoning + Acting. An agent loop where the model alternates between reasoning about its next step and issuing a tool action.; See also: Internal AI OS
Routing: Classifying an input and directing it to the right handler or specialist agent based on its content.
Cascade: A fallback chain where a cheaper or faster model tries a request first and a stronger model handles harder cases.
Planner-Executor: A pattern where one stage plans the work, another executes it, and a verifier checks completion before the system advances.
Self-Critique: Having a model review and improve its own output against explicit criteria before returning it.
Evaluator-Optimizer: A loop that runs a generator against an evaluator, iterating until the output passes a quality bar.
Structured Output: Constraining a model's output to valid JSON that matches a defined schema, rather than free-form text.
Guardrails: Controls that prevent a model from taking unsafe, unintended, or out-of-policy actions.
Squish: A memory system that gives AI agents persistent, searchable memory across sessions.; See also: Agent Memory with Squish
Evals: Automated tests that score model outputs against expected results to measure and track quality over time.
Human-in-the-loop: A workflow where a human approves or handles steps an agent cannot safely complete on its own.; See also: AI Escalation Queue
pgvector: A PostgreSQL extension for storing and querying vector embeddings, used to power semantic search in RAG systems.; See also: RAG Over Company Documents
SKILL.md: A file format for packaging a reusable agent capability as markdown instructions plus any supporting assets.
Build Prompt: A detailed prompt pasted into an AI coding tool such as Claude Code, OpenCode, or Codex to scaffold a working system.
Retrieval: The step in a RAG pipeline that finds the most relevant chunks from a knowledge base for a given query.; See also: RAG Over Company Documents
Lead Enrichment: Augmenting a lead record with additional context such as company, role, and likely pain points.; See also: Lead Scraper to CRM
CRM Automation: Automatically enriching, scoring, tagging, and routing CRM leads using AI so sales teams spend less time on data entry.; See also: CRM Automation Pipeline
Speculative Decoding: An inference optimization where a small draft model proposes tokens and a larger model verifies them in parallel.
Knowledge Distillation: Training a smaller model to imitate a larger one, reducing cost while preserving much of its capability.