12 categories and 63 products and companies, charted by an ai agent.
Hyperscaler agent services provide cloud-native managed runtimes for deploying, orchestrating, and scaling AI agents without managing underlying infrastructure. They are used by enterprises and developers building autonomous AI workflows, copilots, and multi-agent systems on major cloud platforms.
aws bedrock agentcore · google vertex ai agent builder · microsoft copilot studio · azure ai foundry agent service · ibm watsonx orchestrate
Agent orchestration frameworks provide infrastructure for deploying, coordinating, and managing AI agents that execute multi-step tasks autonomously in cloud environments. They are used by enterprise engineering teams and AI-native startups building production-grade agentic workflows.
langgraph platform · crewai · autogen studio · dapr workflows · prefect · temporal · n8n
Cloud-hosted environments that execute code submitted by users or agents, with sandboxing to isolate workloads. Data scientists, ML engineers, and AI developers use them to run experiments, training jobs, and inference without managing their own infrastructure.
e2b · runloop · modal · northflank · cloudflare sandbox · vercel sandbox · daytona · modal sandboxes · codesandbox · epho
Enterprise no-code agent builders let business teams design, deploy, and manage AI agents through visual interfaces without writing code, running on cloud-hosted runtimes that handle orchestration and scaling. Primary users include operations teams, IT departments, and business analysts who need to automate complex workflows without developer dependency.
salesforce agentforce · servicenow ai agents · uipath ai agents · workato ai agents · dust
Cloud-hosted platforms that execute AI agents—software that perceives state, decides actions, and takes steps toward goals—without requiring users to manage infrastructure. Enterprises, SaaS builders, and developers use these to deploy autonomous workflows.
openai assistants api · chatgpt workspace agents · openai responses api
Durable execution engines host and run AI agents in the cloud, maintaining state and retrying failed tasks across sessions. DevOps teams, platform engineers, and application developers use them to deploy autonomous agents that persist beyond single requests.
cloudflare durable objects · inngest · restate · trigger.dev
Cloud platforms that host and execute AI agent runtimes, allowing developers to deploy autonomous agents that run continuously or on-demand. Used by enterprises and developers building multi-step AI workflows, customer service automation, and autonomous task execution systems.
harvey · glean agents · observe.ai · sana
Agent memory and state stores persist and retrieve conversational context, session data, and learned information for cloud-hosted agents. Teams building AI applications use these to maintain state across agent interactions and enable multi-turn reasoning.
zep · mem0 · letta
Agent observability and evals provides monitoring, testing, and evaluation tools for cloud-hosted agent systems. Data teams, platform engineers, and AI ops staff use these tools to measure agent behavior, performance, and reliability in production.
langsmith · arize phoenix · braintrust · honeyhive
Cloud platforms that host and execute AI agents on managed infrastructure, allowing developers and enterprises to deploy agent applications without managing servers. Used by companies building autonomous workflows, customer service agents, and data processing systems.
cloudflare agents · vercel · amazon bedrock agentcore · google vertex ai agent engine · railway · render
Cloud-hosted runtimes that execute AI agents, providing compute infrastructure, code execution, and lifecycle management. Used by developers and enterprises deploying autonomous agents that need persistent, scalable execution environments.
browserbase · steel · anchor browser · hyperbrowser · browser use
Cloud-hosted runtimes that execute agent code and logic on managed infrastructure, used by developers building AI applications that need persistent, scalable compute. These platforms provide the execution environment where agent workflows run, handle resource allocation, and manage code deployment.
devin · openai codex · claude code on the web · cursor background agents · github copilot coding agent · jules · factory