Glossary A - C

Controlled experimentation method that compares two variants of an element to determine which produces better measurable results with statistical significance.

Open protocol developed by Google for standardized communication between AI agents, enabling agents from different providers to collaborate interoperably.

Multiple Claude instances working in parallel on a shared codebase, each with its own context window, capable of communicating with each other and completing complex tasks without active human intervention.

Professional who designs, implements and optimizes AI agent workflows for software development. The most in-demand tech role of 2026 with salaries 40% higher than Senior Software Engineers.

Development paradigm where AI agents execute complete coding tasks autonomously, not just assist. Represents the future of software development.

Discipline focused on ensuring AI systems act in accordance with human intentions and values, preventing undesired or harmful behaviors.

Using AI tools to automatically review code for bugs, security vulnerabilities, style issues, and performance problems.

Intermediary layer that manages, routes, and controls calls to AI model APIs, providing observability, caching, rate limiting, and failover across providers.

Practice of monitoring, measuring, and diagnosing AI system behavior in production, including response quality, latency, costs, and anomaly detection.

Development practice where a human developer works alongside an AI coding assistant in real-time to suggest code, catch errors, and accelerate development.

Research and practice field dedicated to preventing AI systems from causing unintended harm, including alignment techniques, risk assessment, and control mechanisms.

Design of how agents collaborate and coordinate in complex pipelines. Defines roles, communication protocols, and agent handoffs.

Terminal-based AI pair programming tool that enables editing code in Git repositories through natural language conversations with language models.

AWS managed relational database compatible with MySQL and PostgreSQL, delivering up to 5x performance and a serverless option.

Fully managed AWS service providing unified API access to foundation models from multiple providers without managing ML infrastructure.

AWS content delivery network (CDN) service that delivers data, videos, and applications with low latency globally.

AWS monitoring and observability service that collects metrics, logs, and alarms for cloud resources and applications.

AWS identity service providing authentication, authorization, and user management for web and mobile applications with external identity provider support.

High-performance block storage service designed for use with EC2 instances on AWS.

AWS cloud computing service that provides resizable virtual servers to run applications at any scale.

AWS fully managed container orchestration service that runs Docker containers at scale without managing underlying infrastructure.

AWS managed in-memory caching service compatible with Redis and Memcached to accelerate application performance.

Serverless event bus that connects applications using data from your own services, SaaS applications, and AWS services.

AWS managed relational database service supporting multiple engines including PostgreSQL, MySQL, and SQL Server.

Scalable, highly available AWS DNS service that translates domain names to IP addresses and provides intelligent traffic routing.

AWS object storage service designed to store and retrieve any amount of data with high durability and availability.

Scalable AWS email service for transactional and marketing email delivery with high deliverability and competitive pay-per-use pricing.

Fully managed pub/sub messaging service for distributing notifications to multiple subscribers simultaneously.

Fully managed message queue service that enables decoupling and scaling of microservices, distributed systems, and serverless applications.

Isolated virtual network within AWS that provides full control over network configuration, subnets, and cloud resource security.

Service that acts as a single entry point for API consumers, handling routing, authentication, rate limiting, and request transformation.

Strategy for managing API changes without breaking compatibility with existing clients, enabling controlled evolution of the public interface.

Metric representing the annualized recurring revenue of a subscription business, used to evaluate the scale and growth trajectory of SaaS companies.

Core component of the Transformer architecture that enables the model to weigh the relevance of each part of the input when generating each output token.

Ability to automatically adjust compute resources based on actual demand, scaling up or down according to defined metrics.

Automatic detection and correction of problems

Microsoft multi-agent conversation framework where agents interact through messages with human-in-the-loop support.

AWS platform for building and deploying full-stack web and mobile applications with integrated serverless backend, authentication, and hosting.

AWS service for setting custom cost and usage budgets with automatic alerts when spending approaches or exceeds defined thresholds.

AWS infrastructure-as-code framework that lets you define cloud resources using programming languages like TypeScript, Python, or Java instead of YAML or JSON.

AWS continuous delivery service that automates build, test, and deploy pipelines for applications with every source code change.

AWS tool for visualizing, analyzing, and managing cloud service costs and usage over time with detailed reports and forecasting.

Serverless compute engine for containers that eliminates the need to manage servers or EC2 instance clusters.

Serverless AWS data integration service for discovering, preparing, and transforming data across different sources through managed ETL pipelines.

AWS identity and access management service that provides granular control over who can do what with cloud resources.

Amazon serverless compute service that runs code in response to events with automatic scaling and pay-per-execution pricing.

Flexible AWS pricing model that offers discounts in exchange for a per-hour compute spend commitment over 1 or 3 years.

AWS service for securely managing, rotating, and retrieving credentials, API keys, and other secrets required by applications.

Serverless AWS service for orchestrating complex workflows through visual state machines that coordinate multiple services reliably.

AWS web application firewall that protects against common exploits like SQL injection, cross-site scripting, and DDoS attacks.

AWS best practices framework for designing and operating reliable, secure, efficient, and cost-effective cloud systems organized across six pillars.

AWS distributed tracing service that lets you analyze and debug production applications by visualizing the complete request journey across multiple services.

Deployment strategy using two identical production environments to achieve zero-downtime deployments and instant rollback capability.

Financial analysis determining the equilibrium point where self-hosted infrastructure becomes more economical than cloud/API. For AI inference, typical break-even is 4 months with utilization >20%.

The moment when revenues or savings generated by an investment equal the initial cost, with no losses or gains.

Metric calculating the total cost of acquiring a new customer, including marketing and sales, fundamental for evaluating growth sustainability.

Deployment strategy that releases a new version to a small subset of users before rolling it out to the entire base, minimizing the risk of errors.

Difference between capital investments in assets (CapEx) and recurring operational expenses (OpEx), critical for cloud decisions.

Geographically distributed network of servers that delivers cached content from the closest location to the user, reducing latency and absorbing traffic spikes.

Prompting technique where AI models break down complex problems into intermediate reasoning steps for improved accuracy.

Discipline introducing controlled failures into production systems to discover weaknesses before they cause real incidents, improving system resilience.

Metric measuring the percentage of customers who stop using a product or cancel their subscription in a given period, a key indicator of business health.

Test and deployment automation

Resilience pattern that prevents cascading failures in distributed systems by stopping requests to a repeatedly failing service and allowing recovery time.

Anthropic agentic coding tool that operates in the terminal, capable of autonomously planning, writing, testing, and debugging code.

Software architecture pattern by Robert C. Martin that enforces separation of concerns through concentric layers, isolating business logic from frameworks and databases.

AI coding agent for VS Code that can execute complete development tasks autonomously, including file editing, command execution, and web browsing.

Surface-level indicator in source code that suggests a deeper design problem, without necessarily being a bug.

AI coding assistant developed by Sourcegraph that leverages enterprise-scale code search to provide full codebase context in its responses.

Alignment method developed by Anthropic where an AI model self-critiques and self-corrects following a set of constitutional principles.

Lightweight, isolated runtime environment that packages an application with all its dependencies, ensuring consistency between development and production environments.

Optimization of how AI agents access and process codebase information. Reduces inference costs 60-80% and improves accuracy.

Maximum amount of tokens an LLM can process in a single request, determining how much information it can "remember" when generating responses. Claude 4.5 offers up to 1M tokens, GPT-5 up to 400K.

Pattern that separates read and write operations into different models, optimizing each path independently for performance and scalability.

Open-source multi-agent framework based on role assignment where each agent has a defined role, goal, and backstory.

Metric measuring immediate customer satisfaction with a specific interaction, product, or service through point-in-time numerical scale surveys.

CTO as a Service is an external technical leadership model for companies without an internal CTO. Learn what it includes, when it makes sense and what to expect.

Code editor with integrated AI agents. Reached $2.5B valuation in 18 months implementing multi-agent orchestration.