The Hokai Glossary
A comprehensive AI glossary for the Hokai documentation hub. Terms are alphabetically organized. Each entry has a one-line definition and a short explanation.
A
API — Application Programming Interface. A way for software to talk to other software. AI tools often expose APIs for integration. You send a request; you get a response. APIs enable automation and custom workflows.
Agent — AI software that can take autonomous actions, not just generate text. Agents use tools: APIs, search, code execution. They act; chatbots respond. See >What Is an AI Agent?.
API Credits — Prepaid units for API usage. Buy credits; each request consumes some. Common for image and specialized APIs. Different from per-token pricing.
B
Bias — Systematic skew in AI output that disadvantages certain groups. Can come from training data or design. Mitigated by auditing, diverse data, and human review. Important for hiring, lending, and support.
C
Chatbot — AI that converses with users via text or voice. Responds to prompts. Does not take actions. Used for support, sales, and general Q&A.
Context Window — The amount of text (in tokens) an AI model can process at once. Larger context means longer documents and conversations. Ranges from 8K to 1M+ tokens depending on model.
Copilot — AI that works alongside you in real time, suggesting and assisting. Embedded in tools (IDE, doc, spreadsheet). Assists; does not act autonomously. See >What Is an AI Copilot?.
D
Data Residency — Where data is stored geographically. Some regulations require data to stay in a specific region (e.g., EU). Tools may offer region choice for compliance.
DPA — Data Processing Agreement. Contract between you and a vendor that processes personal data. Required under GDPR and similar laws. Specifies how data is handled, stored, and protected.
E
Embeddings — Numerical representations of text (or images) that capture meaning. Similar content has similar vectors. Enable semantic search and RAG. See >Embeddings and Vector Databases.
EU AI Act — European Union regulation for AI. Risk-based. Transparency obligations (Article 50) from August 2026. Affects providers and deployers. Non-EU companies serving EU users are in scope.
Evaluation Scorecard — A template for scoring AI tools on capability, pricing, integration, and other factors. Used to compare and decide. See >Evaluation Scorecard.
F
Fine-Tuning — Training a pre-trained model on your data to adapt its behavior. Changes style, format, or domain knowledge. More complex than prompts or RAG. See >Fine-Tuning vs. Pre-Training.
Foundation Model — A large AI model trained on broad data. Serves as a base for many applications. GPT, Claude, Gemini, Llama are foundation models. See >What Is a Foundation Model?.
Freemium — Free tier with paid upgrades. Try for free; pay for more. Common for SaaS and AI tools.
G
GDPR — General Data Protection Regulation. EU law governing personal data. Consent, access, deletion, and data processing agreements. Applies when you process EU residents' data.
Generative AI — AI that generates new content: text, images, audio, video. Trained on existing data; produces novel output. LLMs and image generators are generative.
H
Hallucination — When AI generates confident but false information. Invented facts, fake citations, plausible nonsense. Reduced by RAG, grounding, and verification. See >AI Hallucinations Explained.
I
Inference — Running a trained model to produce output. The opposite of training. When you send a prompt and get a response, that is inference.
Integration — How well tools connect to each other. Native integrations, APIs, webhooks. Good integration reduces manual work and enables automation.
L
LLM — Large Language Model. A foundation model trained on text. Can generate, summarize, translate, and reason. GPT, Claude, and Gemini are LLMs.
Low-Code — Development with minimal hand-coding. Visual builders, drag-and-drop. Faster than full code; more flexible than no-code.
M
MCP — Model Context Protocol. Anthropic's open protocol for connecting AI to external tools and data. Servers expose tools; clients let models call them. See >What Is MCP?.
Model Directory — Hokai's searchable database of AI tools. Categories, filters, pricing, and profiles. Use to explore and compare. See >Navigating the Directory.
Multi-Modal — AI that processes and generates multiple content types: text, image, audio, video. GPT-4o, Claude, and Gemini are multi-modal. See >What Is Multi-Modal AI?.
My Stack — Hokai feature for managing your AI tools. Track what you use, cost, and health. Add, remove, optimize. See >My Stack Overview.
N
No-Code — Building without writing code. Configurable tools, visual workflows. Good for non-developers. Less flexible than code.
P
Parameters — Values the model learns during training. More parameters often mean more capability. Billions for modern LLMs. Not the same as context window.
Per-Seat — Pricing per user. $X per person per month. Common for team tools. Cost scales with headcount.
PII — Personally Identifiable Information. Data that can identify a person. Name, email, ID numbers. Protected by GDPR, CCPA, and similar laws.
Playbook — Hokai's educational guides and step-by-step content. How-to and strategy. Part of Content and Community.
Pre-Training — Initial training of a foundation model on massive data. Done by model providers. You do not pre-train; you use or fine-tune pre-trained models.
Prompt — The input you give an AI. Text (or other modalities) that instructs or questions. Quality of prompt affects quality of output.
Prompt Engineering — Crafting prompts to get better outputs. Specificity, examples, roles, constraints. See >What Is Prompt Engineering?.
Pulse — Hokai's feed for AI tool updates. Price changes, new features, deals. Stay informed on your stack. See >The Pulse News Feed.
R
RAG — Retrieval-Augmented Generation. Give AI access to your documents before answering. Reduces hallucinations. Powers knowledge bases and Q&A. See >What Is RAG?.
Rate Limit — Cap on requests per minute or day. Free tiers often have strict limits. Paid tiers relax them. Hitting the limit blocks further requests.
RLHF — Reinforcement Learning from Human Feedback. Training method where humans rate outputs; model learns to produce higher-rated responses. Used by OpenAI, Anthropic, and others.
S
SaaS — Software as a Service. Cloud-hosted, subscription-based. Most AI tools are SaaS.
Smart Match — Hokai's conversational AI for stack recommendations. Describe your needs; get a Strategy Brief and ranked stack. Persona: Hok. See >How Smart Match Works.
SOC 2 — Security audit framework. Vendors get audited on security, availability, and related controls. Type II means operating effectiveness. Often required by enterprise buyers.
Strategy Brief — Output from Smart Match. Summarizes your context and recommends a stack. Includes rationale and next steps.
T
Temperature — Setting that controls randomness in model output. Low (0) = deterministic. High (1) = creative, variable. Use low for factual tasks; higher for creative.
Tokens — Chunks of text the model processes. Roughly 4 characters or 0.75 words in English. API pricing is often per token. Input and output tokens may have different rates.
Transformer — Neural network architecture behind modern LLMs. Attention mechanism. Enables long-range dependencies and scale. GPT and BERT use transformers.
U
Usage-Based — Pay for what you use. No fixed fee. Bills vary by month. Common for APIs and consumption-based tools.
V
Vector Database — Storage for embeddings. Optimized for similarity search. Returns nearest neighbors to a query vector. Powers RAG and recommendations. Examples: Pinecone, Weaviate, Qdrant.
Vibe Coding — Building software through natural language prompts. AI generates code; you iterate. See >What Is Vibe Coding?.
W
Webhook — HTTP endpoint that receives data from external systems. Used for triggers and integrations. Workflow platforms use webhooks to start flows.
Workflow Platform — Tool that connects AI to other software via automated workflows. Triggers, actions, data flow. Examples: Zapier, Make, n8n. See >What Is an AI Workflow Platform?.
The Bottom Line
This glossary defines key AI and Hokai terms. Use it for quick reference. Links to deeper docs where available. Terms and usage evolve; we update regularly.