Symbolic AI - The collection of all methods in artificial intelligence research that are based on high-level symbolic (human-readable) representations of problems, logic and search
Neuro-symbolic AI - A subfield of artificial intelligence that integrates neural methods with symbolic methods to combine the strengths of both approaches, resulting in AI systems that can be trained from raw data while preserving explainability and explicit reasoning
AlphaGeometry - A neuro-symbolic system made up of a neural language model and a symbolic deduction engine, which work together to find proofs for complex geometry theorems
Generative AI - A subset of artificial intelligence that uses generative models to produce text, images, videos, or other forms of data
Causal AI - A technique in artificial intelligence that builds a causal model and can thereby make inferences using causality rather than just correlation
Core Concepts
Embedding - A representation learning technique that maps complex, high-dimensional data into a lower-dimensional vector space of numerical vectors
Transfer learning - A machine learning technique where knowledge gained from one task is reapplied to improve performance on a different but related task
Mathematical model - An abstract description of a concrete system using mathematical concepts and language
Mathematical optimization - The selection of a best element, with regard to some criteria, from some set of available alternatives
Supervised learning - A paradigm in machine learning where algorithms learn from labeled data
Classification - The problem of identifying which of a set of categories (sub-populations) a new observation belongs to, on the basis of a training set of data containing observations
Logistic regression - A statistical model that models the probability of an event taking place by having the log-odds for the event be a linear combination of one or more independent variables
Support vector machine - A set of supervised learning models with associated learning algorithms that analyze data for classification and regression analysis
Naive Bayes classifier - A family of simple probabilistic classifiers based on applying Bayes' theorem with strong (naive) independence assumptions between the features
Decision tree learning - A method using a decision tree as a predictive model to go from observations about an item to conclusions about its target value
Ensemble learning - A method using multiple learning algorithms to obtain better predictive performance than could be obtained from any of the constituent learning algorithms alone
Random forest - An ensemble learning method for classification, regression and other tasks that operates by constructing a multitude of decision trees at training time
ROC curve - A graphical plot that illustrates the diagnostic ability of a binary classifier system as its discrimination threshold is varied
Regression - A set of statistical processes for estimating the relationships between a dependent variable and one or more independent variables
Ordinary least squares - A type of linear least squares method for choosing the unknown parameters in a linear regression model
ARIMA model - A generalization of an autoregressive moving average (ARMA) model, fitted to time series data either to better understand the data or to predict future points in the series
Unsupervised learning - A type of machine learning in which models are trained using unlabeled dataset and are allowed to act on that data without previous training
Principal component analysis - A linear dimensionality reduction technique with applications in exploratory data analysis, visualization and data preprocessing
K-means clustering - A method of vector quantization that aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest mean
Reinforcement learning - An area of machine learning concerned with how intelligent agents ought to take actions in an environment in order to maximize the notion of cumulative reward
Markov chain - A stochastic process that describes a sequence of events where the probability of each event depends only on the state attained in the previous event
Markov decision process - A mathematical framework for modeling decision making in situations where outcomes are partly random and partly under the control of a decision maker
Hidden Markov model - A statistical Markov model where the system being modeled is assumed to be a Markov process with unobserved (hidden) states
Multi-armed bandit - A problem in which a fixed limited set of resources must be allocated between competing (alternative) choices in a way that maximizes their expected gain
Value function - A function used in mathematical optimization and reinforcement learning that assigns a measure of desirability to states or actions
Hyperparameter - A parameter whose value is used to control the learning process
Hyperparameter optimization - The problem of choosing a set of optimal hyperparameters for a learning algorithm
Early stopping - A form of regularization used to avoid overfitting when training a learner with an iterative method, such as gradient descent
Cross-validation - Any of various similar model validation techniques for assessing how the results of a statistical analysis will generalize to an independent data set
Anomaly detection - The identification of rare items, events or observations which raise suspicions by differing significantly from the majority of the data
One-class classification - The technique trying to identify objects of a specific class amongst all objects, by primarily learning from a training set containing only the objects of that class
Recommender system - An information filtering system that seeks to predict the 'rating' or 'preference' a user would give to an item
ML.NET - An open-source, cross-platform machine learning framework for .NET developers
Crab - A Python library for building recommender systems
mlxtend - A Python library of useful tools for the day-to-day data science tasks
Prophet - A forecasting procedure for time series data that is fast and provides completely automated forecasts
Azure Machine Learning - An enterprise-grade machine learning service to build and deploy models faster
Amazon SageMaker - The service to build, train, and deploy machine learning (ML) models for any use case with fully managed infrastructure, tools, and workflows
Neural network - The computational models used in machine learning for finding patterns in data
Tensor - The mathematical objects represented as multidimensional arrays used in machine learning
Activation Functions
Sigmoid function - A mathematical function having a characteristic 'S'-shaped curve or sigmoid curve
Softmax function - A function that converts a vector of K real numbers into a probability distribution of K possible outcomes
Backpropagation - A widely used algorithm for training feedforward neural networks
Autoencoder - A type of artificial neural network used to learn efficient codings of unlabeled data (unsupervised learning)
Vanishing gradient problem - The difficulty encountered when training artificial neural networks with gradient-based learning methods and backpropagation, where gradients shrink as they back-propagate
Deep Learning Concepts & Training
Deep Learning - A part of a broader family of machine learning methods based on artificial neural networks with representation learning
Stochastic gradient descent - An iterative method for optimizing an objective function with suitable smoothness properties
Fine tuning - An approach to transfer learning in which the weights of a pre-trained model are trained on new data
LoRA (machine learning) - A parameter-efficient fine-tuning technique for adapting pre-trained models to specific tasks with significantly fewer computational resources
Recurrent neural network - A class of artificial neural networks where connections between nodes can create cycles, allowing output from some nodes to affect subsequent input to the same nodes
LSTM - An artificial neural network used in the fields of artificial intelligence and deep learning, distinguished by feedback connections
Attention - A technique in the context of neural networks that mimics cognitive attention, enhancing the important parts of the input data and fading out the rest
FlashAttention - A fast and memory-efficient exact attention mechanism
Transformer - A deep learning architecture based on the multi-head attention mechanism
Deep Learning, MIT Press - The textbook intended to help students and practitioners enter the field of machine learning in general and deep learning in particular
Morphology - The study of words, how they are formed, and their relationship to other words in the same language
Syntax - A linguistic field that is the study of how words and morphemes combine to form larger units like phrases and sentences
Semantics - The study of linguistic meaning, examining how words acquire meaning and how complex expressions derive meaning from their constituent parts
Symbol grounding problem - The challenge of connecting abstract symbols to the real-world objects or concepts they represent
Okapi BM25 - A ranking function used by search engines to estimate the relevance of documents to a given search query based on a probabilistic retrieval framework
Levenshtein distance - A string metric for measuring the difference between two sequences by counting the minimum number of single-character edits required to change one into the other
n-gram - A sequence of n adjacent symbols in a particular order, used in fields like natural language processing and computational biology
tf-idf (term frequency-inverse document frequency) - A statistical measure used in information retrieval to evaluate the importance of a word in a document relative to a collection or corpus, accounting for its general frequency
Word embedding - A representation of a word in natural language processing, typically a real-valued vector that encodes its meaning such that words closer in vector space are expected to be similar in meaning
Word2vec - A technique in natural language processing for obtaining vector representations of words that capture information about their meaning based on surrounding words
fastText - Library for efficient text classification and representation learning
OpenCV - An open source computer vision and machine learning software library
GoCV - A package for the Go programming language with bindings for OpenCV 4
Tesseract OCR - An open source text recognition (OCR) Engine
gosseract OCR - A Go package for OCR (Optical Character Recognition), by using Tesseract C++ library
EasyOCR - A ready-to-use OCR with 80+ supported languages and all popular writing scripts
OCRmyPDF - A tool to add a searchable OCR text layer to PDF files
Open Models
LLaVA - A novel end-to-end trained large multimodal model that combines a vision encoder and Vicuna for general-purpose visual and language understanding, achieving impressive chat capabilities mimicking spirits of the multimodal GPT-4 and setting a new state-of-the-art accuracy on Science QA
Diffusion model - A class of latent variable generative models in machine learning that learn to generate new data by reversing a gradual noising process
Multimodal learning - A deep learning approach that combines and processes diverse data types such as text, audio, images, or video for a more holistic understanding of complex information
Anthropic - The API providing access to Anthropic's Claude models
OpenAI - The platform for building applications with OpenAI's models
DeepSeek - An AI model research and development company that focuses on building advanced large language models and artificial intelligence infrastructure
Kimi - An AI assistant platform by Moonshot AI featuring the K2 model with long-context capabilities, designed for coding assistance, deep research, and multi-agent workflows
Gemini Developer APIs - The API that gives you access to the latest Gemini models from Google
Go OpenAI - The Go client libraries for OpenAI API
Google Gen AI SDK - The Python SDK for Google's generative AI models
Instructor - A Python library designed to extract structured, validated data from Large Language Models (LLMs)
OmniAI - A minimalist library for interfacing with LLMs
Outlines - A library that guarantees structured outputs during generation directly from any large language model by enforcing compliance with JSON Schema, regular expressions, or context-free grammars
Retrieval-augmented generation (RAG) - A technique that enables large language models to retrieve and incorporate new information from external data sources
dsRAG - A high-performance retrieval engine for unstructured data
GraphRAG - A data pipeline and transformation suite that is designed to extract meaningful, structured data from unstructured text using the power of LLMs
Prompt Engineering Guide - A comprehensive resource for learning and applying prompt engineering techniques to effectively utilize large language models and build AI agents
CRAFT framework - A structured method for crafting clear and precise AI prompts by defining context, role, action, format, and tone
LiteLLM - A Python SDK and AI Gateway (Proxy) that allows users to call over 100 Large Language Models (LLMs) using a unified OpenAI input/output format
RedCandle - A Ruby gem for running state-of-the-art language models locally (via Rust's Candle)
Unsloth AI - A platform providing tools and services for easily fine-tuning and training Large Language Models (LLMs) to achieve faster and more efficient AI training
LLM - A CLI utility and Python library for interacting with Large Language Models
lootbox - A CLI which is inspired by "Code Mode" - LLMs write TypeScript code to call APIs rather than using tool invocation
Gradio - The fastest way to demo your machine learning model with a friendly web interface so that anyone can use it, anywhere
ARC-AGI - The benchmark measuring progress toward artificial general intelligence by testing skill-acquisition efficiency on novel tasks that are intuitive for humans but challenging for AI systems
OSWorld - A scalable, real computer environment for evaluating multimodal agents on 369 open-ended tasks spanning web applications, desktop software, and cross-application workflows across Ubuntu, Windows, and macOS
FrontierMath - An AI benchmark consisting of extremely challenging mathematical problems, including open research problems authored by expert mathematicians, ranging from undergraduate to research-level difficulty
MRCR v2 - A multi-round context recall benchmark that evaluates LLMs on their ability to retrieve and use information from extended multi-turn conversation histories
Artificial Analysis - An independent analysis of AI models and API providers, helping users understand the AI landscape
Arena - A platform designed for benchmarking and comparing various AI models, including both large language models (LLMs) and vision-language models (VLMs)
AGENTS.md - An open standard for defining and running AI agents
Agent Skills - A simple, open format for giving agents new capabilities and expertise
Model Context Protocol (MCP) - An open-source standard for connecting AI applications to external systems, enabling them to access data sources, tools, and workflows
A2A Protocol - A protocol for enabling bidirectional communication between web applications and AI agents
Agent Name Service (ANS) - A secure, DNS-inspired framework for AI agent discovery that leverages Public Key Infrastructure (PKI) for identity verification, structured JSON schemas for communication, and a protocol adapter layer supporting A2A, MCP, and ACP protocols
GitAgent - A framework-agnostic standard that allows you to define an AI agent as a set of version-controlled files within a Git repository
ReAct Prompting - A prompting technique synergizing reasoning and acting in language models
Reason, Act, Thought, Observation
Recursive Language Models - An inference strategy where language models (LMs) can decompose and recursively interact with input context of unbounded length
Effective Context Engineering for AI Agents - The set of strategies for curating and maintaining the optimal set of tokens (information) during LLM inference, including all the other information that may land there outside of the prompts
Agno - A multi-agent framework, runtime and control plane
crewAI - An open-source, multi-agent orchestration framework that empowers developers to orchestrate high-performing AI agents with ease and scale
Deep Agents - The easiest way to start building agents and applications powered by LLMs—with built-in capabilities for task planning, file systems for context management, subagent-spawning, and long-term memory
Deep Agents CLI - A terminal coding agent built on the Deep Agents SDK
LangGraph - A library for building stateful, multi-actor applications with LLMs by creating cyclic graphs for agent runtimes
Mastra - An all-in-one, open-source TypeScript framework for building, iterating, and deploying AI agents with built-in support for workflows, RAG, memory, and observability
Microsoft Agent Framework - A resource for building robust, future-proof Agentic AI solutions that evolve with technological advancements
Microsoft 365 Agents SDK - A comprehensive framework for building full-stack, multi-channel agents that operate seamlessly across Microsoft 365 Copilot, Teams, third-party platforms, custom applications, and websites
Application Frameworks
Chainlit - An open-source Python package to build production ready Conversational AI
DSPy - A declarative framework for building modular AI software that allows for fast iteration on structured code and offers algorithms to compile AI programs into effective prompts and weights for language models
Genkit - The AI framework for building full-stack applications with integrated support for agents, RAG, and tool use
LangChain - A framework for developing applications powered by large language models
LlamaIndex - A developer-first agent framework that rapidly accelerates time-to-production of GenAI applications with trusted low and high-level abstractions
PydanticAI - A Python agent framework for building production-grade applications with Generative AI, emphasizing type safety and structured outputs
Agent Package Manager (APM) - An open-source dependency manager for AI agents that provides a single source of truth for skills, prompts, and instructions
Claude Agent SDK - The Agent SDK gives you the same tools, agent loop, and context management that power Claude Code, programmable in Python and TypeScript
Claude Agent Skills - The modular capabilities that extend an agent's functionality by packaging instructions, metadata, and optional resources
Fantasy - A Go library for building AI agents with multiple providers and models through a single API
Microsoft 365 Agents Toolkit - A suite of tools for building enterprise-ready agents and apps that work across Microsoft 365 Copilot, Teams, Office, web, and other third-party messaging channels
OpenAI Agents SDK - A library for building agentic applications where models can use additional context and tools, hand off to specialized agents, and stream results
Foundry Agent Service - A platform to securely design, deploy, and scale AI agents with governance and observability for enterprise transformation
Moltbook - A social network for AI agents where AI agents share, discuss, and upvote
Interoperability
FastMCP - A Pythonic framework for building Model Context Protocol (MCP) servers and clients
Pre-built Agents & Collections
agency-agents - A growing collection of meticulously crafted AI agent personalities designed to act as specialized experts with unique voices, proven workflows, and measurable deliverables
Memory Systems
Mem0 - An AI memory layer for LLM applications that aims to provide personalized AI experiences
Graphiti - An open-source Python framework for building temporally-aware context graphs
Search & Data Extraction
Firecrawl - An API service that takes a URL, crawls it, and converts it into clean markdown or structured data
Tavily Search - A search engine optimized for LLMs, aimed at efficient, quick and persistent search results
SWIRL AI Search - A Federated AI Search solution that connects to over 100 enterprise platforms, providing real-time visibility across data and document silos without requiring costly data transformations or migrations
Security Tools
skill-scanner - A best-effort security scanner for AI Agent Skills that detects prompt injection, data exfiltration, and malicious code patterns
ClawShell - The Runtime Security Layer for OpenClaw, the essential safety harness for PII & sensitive credentials protection
DVC - Open-source Data Version Control for machine learning projects
CML - An open-source tool for implementing continuous integration & delivery (CI/CD) in machine learning projects
MLFlow - An open source platform to manage the ML lifecycle, including experimentation, reproducibility, deployment, and a central model registry
KubeFlow - The Machine Learning Toolkit for Kubernetes, dedicated to making deployments of ML workflows on Kubernetes simple, portable and scalable
Managed MLOps Platforms
Microsoft Foundry - A unified, interoperable platform for building, optimizing, and governing AI apps and agents that understand business context and deliver business impact
Vertex AI - A machine learning (ML) platform for training and deploying ML models and AI applications
Nebius - A specialized AI cloud platform offering purpose-built infrastructure for AI and machine learning workloads
Amazon Bedrock - A fully managed service offering a choice of high-performing foundation models
Amazon Bedrock Agents - A service that uses the reasoning of foundation models, APIs, and data to break down user requests, gather relevant information, and efficiently complete tasks
Agent Observability
Mission Control - A centralized operational control plane to manage, monitor, and coordinate fleets of AI agents
LLM Observability & Evals
Langfuse - An open source LLM engineering platform providing traces, evals, prompt management, and metrics to debug and improve LLM applications
OpenLIT - An open-source, OpenTelemetry-native tool for LLM and GenAI observability
LangSmith - A unified DevOps platform for developing, collaborating, testing, deploying, and monitoring LLM applications
Helicone - An open-source LLM observability platform built for developers to monitor and optimize their generative AI applications
Arize Phoenix - An open-source AI observability and evaluation platform for LLMs
Braintrust - The enterprise AI platform that provides an evaluation and observability platform for developers building with LLMs