Startup Super School
53 resources

AI for Startups

Everything founders, builders, and operators need to understand AI — from foundational concepts to practical guides on building, deploying, and scaling AI products.

Deep Dives

View all articles →

How-To Guides

View all guides →

Frameworks & Concepts

View all concepts →

AI Glossary

View full glossary →
Agentic Workflow
A multi-step AI process where a model autonomously plans, uses tools, and executes tasks without human input at each step.
AI Agent
An AI agent is a system that uses an LLM to autonomously plan, make decisions, use tools, and take actions to complete a goal.
AI Alignment
The challenge of ensuring AI systems behave in ways that match human intentions, values, and goals.
AI Hallucination
When an AI model generates confident-sounding but factually incorrect or fabricated information.
AI Wrapper
An AI wrapper is a product built on top of a foundation model API with a custom UI, workflow, or niche focus, rather than novel AI model development.
Claude Code
Anthropic's agentic CLI tool that gives Claude full access to your codebase, enabling multi-file edits, terminal commands, and autonomous coding tasks.
Context Window
The maximum amount of text an LLM can process in a single interaction - inputs plus outputs combined.
DeepSeek
A Chinese AI lab and open-source model family that trained frontier-level LLMs at a fraction of Western competitors' reported costs.
Embedding
A numerical vector that represents the meaning of text, enabling AI to compare and retrieve semantically similar content.
Fine-Tuning
Fine-tuning adapts a pre-trained LLM to a specific task or domain by continuing training on a curated dataset of examples.
Foundation Model
A foundation model is a large AI model trained on broad data at scale, designed to be adapted to many downstream tasks rather than one specific use case.
Inference
Inference is the process of running a trained AI model on new inputs to generate predictions or outputs, as opposed to training the model on data.
Large Language Model (LLM)
An LLM is a deep learning model trained on massive text datasets to generate, summarize, translate, and reason with human language.
Multimodal AI
AI models that can process and generate multiple types of data - text, images, audio, and video - within a single system.
Open Source AI
AI models whose weights, architecture, and training details are publicly released - enabling free use, modification, and self-hosting.
OpenClaw
An open-source, local-first AI agent platform that integrates with 20+ messaging apps and runs entirely on your own devices.
Prompt Engineering
Prompt engineering is the practice of crafting LLM inputs to reliably produce accurate, useful, and correctly formatted outputs for a given task.
Qwen
Alibaba's open-source large language model family - multilingual, high-performing, and available in sizes from 0.5B to 72B parameters.
Retrieval-Augmented Generation (RAG)
RAG is an AI architecture that combines a retrieval system with an LLM, giving the model access to external knowledge at query time.
Synthetic Data
Artificially generated data that mimics real data - used to train, test, and fine-tune AI models when real data is scarce or private.
Token (AI)
A token is the basic unit of text an LLM processes - roughly 3–4 characters or 0.75 words - used to measure input length, output length, and API cost.
Vector Database
A database optimized for storing and searching vector embeddings - the backbone of AI-powered search and RAG systems.
Vertical AI
Vertical AI is an AI product built for a specific industry or workflow, combining foundation model capabilities with deep domain expertise and proprietary data.
Zero-Shot Learning
An AI model's ability to perform a task it was never explicitly trained on, guided only by a natural language description.