The AI tool directory for founders & agents

Microsoft's framework for multi-agent conversation, with a programming model for agents that talk to each other and tools.

Role-playing autonomous agents that collaborate as a 'crew' with defined roles, goals, and task delegation.

Data framework for connecting LLMs to private data — indexing, retrieval, and agentic RAG over your documents.

Batteries-included agent runtime — built-in memory, knowledge/RAG, and AgentOS, a control plane you run in your own cloud. Formerly Phidata.

Graph-based orchestration for stateful, multi-actor agent workflows with explicit control flow and checkpointing.

Programming — not prompting — language models: compile declarative pipelines into optimized prompts/weights.

OpenAI's lightweight agent framework — a small set of primitives (Agents, Handoffs, Guardrails, Sessions); provider-agnostic via LiteLLM. Evolved from Swarm.

Google's Agent Development Kit — a code-first, model-agnostic toolkit for building, evaluating, and deploying multi-agent systems. Optimized for Gemini.

Type-safe agent framework from the Pydantic team — structured outputs, dependency injection, and model-agnostic agents.

Anthropic's SDK for building agents on Claude — the harness behind Claude Code, with in-process tools, hooks, and an interactive client. Formerly the Claude Code SDK.

AWS's model-driven agent SDK — give it a prompt and tools and let the model plan, call tools, and reflect in a loop. First-class Bedrock, MCP-native, OpenTelemetry tracing.

TypeScript-first SDK for stateful agents where each agent is a Durable Object — embedded SQLite, WebSockets, cron scheduling, and hibernation, running globally on Cloudflare's edge.

A memory layer for AI agents — extracts, stores, and retrieves user/agent facts across sessions.

Stateful agents with long-term memory and self-editing context, evolved from the MemGPT research.

Long-term memory store for agents with a temporal knowledge graph of facts and their validity over time.

Cloud-native vector database built for billion-scale similarity search.

In-process analytical database whose vss extension adds an HNSW vector index — vector search alongside your columnar analytics.

High-performance vector search engine with rich filtering, written in Rust for production-scale retrieval.

Open-source embedding database designed for simplicity — the default vector store for many RAG prototypes.

Vector similarity search inside Postgres — keep embeddings next to your relational data.

Open-source vector database with hybrid search and built-in modules for vectorization and RAG.

Embedded, in-process vector database on the columnar Lance format — versioned, updatable, larger-than-RAM retrieval with no server.

A single-file SQLite extension for vector search — exact brute-force KNN that lives inside the database you already ship.

The reference collection of Model Context Protocol servers — connect agents to files, GitHub, databases, and more.

The fast, Pythonic way to build MCP servers and clients — decorators over boilerplate.

Test-driven prompt and agent development — evals, red-teaming, and side-by-side model comparison from the CLI.

Pytest-like framework for unit-testing LLM outputs with metrics for hallucination, relevancy, and bias.

Evaluation toolkit for RAG pipelines — faithfulness, answer relevancy, and context metrics without ground truth.

Open-source LLM engineering platform — tracing, evals, prompt management, and metrics for agent apps.

Arize's open-source observability for LLM apps — OpenTelemetry-based tracing and evaluation.

Open-source observability for LLM apps via a proxy — logging, caching, and cost tracking with one header.

Durable execution platform — write long-running, failure-resilient agent workflows as ordinary code.

Secure cloud sandboxes for running AI-generated code — the runtime layer for code-executing agents.

Independent search-index API (web, news, images) with an LLM-focused Answers/Grounding endpoint, from a privacy-first crawler.

Neural/semantic web search API purpose-built for AI agents, returning clean results plus full page contents and highlights.

Web-data API to search, scrape, crawl and map sites into clean markdown/structured data for LLMs — one call returns full page content.

Google's official JSON API to query a Programmable Search Engine — a cheap, widely-used web/image search baseline for retrieval.

URL-to-LLM-markdown reader (r.jina.ai) and web search (s.jina.ai) that work with a simple URL prefix — even without an API key.

Privacy-first, ad-free search API from Kagi with high-quality results, plus a FastGPT/Summarizer endpoint for LLMs.

Production-grade web search API for AI that returns sourced, cited answers or raw results to reduce hallucination.

Search API built from the ground up for AI agents — pass an objective and get URLs plus token-relevant compressed excerpts.

Search-grounded LLM API (Sonar models) that returns cited answers with live web search built in, plus a raw Search API.

Real-time SERP API across 20+ engines (Google, Bing, YouTube, Maps, Scholar) with a pay-for-success model and structured JSON.

Real-time SERP scraping API returning structured JSON from Google and 20+ engines (maps, news, shopping, scholar).

Fast, low-cost Google Search API returning structured JSON (organic, knowledge graph, news, images, places) for agents and SEO.

Web search, extract, crawl and map API built specifically for LLMs and AI agents, returning LLM-ready structured results.

Search API for agents unifying web plus 50+ proprietary sources (academic papers, SEC filings, financial data, clinical trials).

Real-time web search, page-contents, and multi-step research APIs for AI apps, with a free keyless MCP endpoint.

Speech-to-text and audio-intelligence API — transcription, diarization, summarization, and LLM-over-audio (LeMUR).

Sonic — ultra-low-latency (~90ms), emotive real-time text-to-speech and voice cloning for live voice agents.

Talking-avatar video API — turn a photo + text/audio into a lip-synced presenter, with real-time WebRTC streaming agents.

Fast, accurate speech-to-text, text-to-speech, and voice-agent APIs built for real-time and batch at scale.

Ultra-realistic text-to-speech, voice cloning, dubbing, and voice-agent APIs for developers and creators.

OpenAudio (S1/S2) text-to-speech and voice cloning — open models plus a hosted pay-as-you-go API.

Audio-infrastructure API for voice products — async and real-time speech-to-text with diarization and translation.

Programmatic AI avatar video generation and video translation/dubbing with lip-sync, via a REST API.

Empathic Voice Interface (EVI) and Octave TTS — emotionally intelligent, context-aware expressive speech.

Low-latency (150–200ms) multilingual text-to-speech with studio-quality voices and instant cloning.

Generative text-to-speech API with 900+ voices, instant voice cloning, and low-latency streaming for conversational AI.

Voice cloning and generative speech API, plus deepfake detection and audio watermarking, billed per second.

Voice-agent API for AI phone calls — build, test, and deploy production call agents with usage-based per-minute pricing.

Enterprise TTS with 200+ realistic voices, sub-200ms latency, and linguistically-aware markup for voice agents.

High-accuracy speech-to-text across 55+ languages, plus real-time transcription and the Flow voice-agent API.

Studio-grade AI lip-sync and visual dubbing API — retime any video's mouth to new audio in any language.

Studio-grade AI avatar video generation API — script-to-video with 200+ avatars, translation, and templates.

Conversational Video Interface (CVI) and Phoenix replica API — real-time, face-to-face AI video agents ('digital twins').

Developer-first platform to build, test, and deploy real-time voice AI agents with pluggable STT/LLM/TTS and telephony.

Email inbox API purpose-built for AI agents — programmatically create real inboxes that can send, receive, and reply in threads.

Enterprise voice-AI platform for phone agents — programmatically dial, converse, and transfer calls end to end.

Multi-channel notification API and orchestration for email, push, SMS, in-app, and chat with a single send call.

Notifications infrastructure for developers — orchestrate email, push, SMS, in-app, and chat from one workflow API.

Email platform for SaaS teams — marketing, lifecycle, and transactional email from one contact record and one API, with an agent-focused REST API.

Developer email API with HTTP + SMTP sending, routing, and inbound email parsing.

Email, calendar, and contacts API — plus Agent Accounts that give an AI agent its own real inbox and calendar in one API call.

Pay-as-you-go SMS, voice, and WhatsApp API with a fast developer onboarding path.

Fast, reliable transactional email API and SMTP known for deliverability and long message history.

Developer-first transactional email API with a clean SDK, React email templates, and instant self-serve API keys.

Enterprise communications APIs: SMS across 13+ channels, voice, email, video, and verification on one super-network.

Carrier-grade programmable messaging, voice, phone numbers, and voice-AI on a private global network.

The market-standard programmable communications platform: SMS, voice, WhatsApp, verification, and phone numbers via API.

High-volume transactional and marketing email API (now part of Twilio), widely used at enterprise scale.

SMS, messaging (WhatsApp/Viber), voice, and verification APIs from Vonage (Ericsson).

Marketplace and cloud runtime for thousands of ready-made scrapers ('Actors') that extract data from social media, maps, e-commerce, and any website.

Web-data platform with residential/datacenter proxies, Unlocker, SERP, Scraper APIs, and prebuilt datasets for large-scale collection.

Spreadsheet-style GTM platform that waterfalls 150+ data providers and AI research (Claygent) to enrich leads and build outbound systems.

Workforce and company data via APIs and bulk datasets (firmographics, employees, funding, technographics) refreshed continuously.

Real-time B2B people and company data via APIs and webhooks (search, enrich, and Watcher signal alerts), built for AI agents.

AI web-extraction and a 10B+ entity Knowledge Graph of the public web (people, companies, articles) queryable via API.

Email finder, verifier, and domain search API to turn names/domains into deliverable B2B contact data.

Enterprise proxy network plus Web Scraper API that returns parsed data from complex and SERP targets at scale.

B2B person and company data API over 900M+ profiles for enrichment, search, and identity resolution.

Web-scraping API that handles headless browsers, JS rendering, and proxy rotation so you get HTML/data without managing infrastructure.

AI research agents that enrich people and companies and surface real-time signals from a single API call, built for GTM teams and autonomous agents.

Scraping API and scraping browser with anti-bot bypass, premium proxies, and JS rendering for hard-to-reach sites.

Cloud browser automation controlled with natural language, handling OAuth/2FA/CAPTCHA for sales, marketing and research agents.

AWS service + SDK for building reliable browser agents that automate production UI workflows, with an IDE extension and AWS deploy path.

Secure cloud browsers for computer-use agents, with strong support for authenticated sessions, MFA/SSO and infinite concurrency.

Enterprise-grade unblocking browser (Scraping Browser) with automatic CAPTCHA/fingerprint handling over a huge residential proxy network.

The popular open-source Python agent that drives a browser from natural language, now with a hosted cloud API and stealth browsers.

Managed fleet of headless cloud browsers with stealth, proxies and live view — the 'AWS for browsers' most agent frameworks run on.

Battle-tested headless browser hosting (Chrome/Chromium) with Playwright/Puppeteer endpoints, BrowserQL and built-in CAPTCHA solving.

Cloud Chrome sessions for agents with stealth fingerprinting, proxy rotation and auto-CAPTCHA, plus the open-source HyperAgent framework.

Unikernel-based cloud browsers with sub-150ms cold starts and a full REST control plane (profiles, proxies, replays, computer-use).

An open-source headless browser written from scratch in Zig for AI/automation — ~11x faster and ~9x lighter than Chrome, CDP-compatible.

Browser platform for AI that turns any web workflow into a reliable, deployable API, with web agents and a perception layer for LLMs.

API-driven virtual desktops (Ubuntu/Windows/browser) for agents — full OS + browser control with an Act SDK for computer-use models.

LLM + computer-vision browser automation that reads pages visually instead of brittle CSS selectors; open-source with a hosted cloud.

Open-source SDK for browser agents (act/extract/observe/agent) that mixes natural-language and code; runs locally or on Browserbase.

Open-source headless browser API purpose-built for AI agents, with token-efficient content extraction and auth/session persistence.

Production inference platform: per-token Model APIs plus dedicated GPU deployments for custom and open models.

The fastest LLM inference in 2026 (1,800+ tokens/sec) on wafer-scale CS-3 chips, for open models like Llama and Qwen.

Free edge AI gateway (caching, rate/spend limits, observability) in front of any provider, plus Workers AI for serverless model inference.

Low-cost pay-per-use API for 50+ top open models via an OpenAI-compatible endpoint, no infra to manage.

Fast serverless inference for open models with serverless, on-demand, and reserved-capacity deployment tiers.

Ultra-fast inference on custom LPU chips (500+ tokens/sec) for open models like Llama, Qwen, and DeepSeek distills.

Open-access GPU and AI cloud with OpenAI-compatible serverless inference for 25+ open models plus on-demand GPU rental.

Open-source Python SDK + proxy server (AI gateway) that calls 100+ LLM APIs in OpenAI format with cost tracking, budgets, and virtual keys.

Serverless GPU/CPU cloud where you define container + hardware in Python code and run per-second-billed inference, batch, and training jobs.

OpenAI-compatible inference and fine-tuning for open models with base/fast checkpoint variants and transparent per-token pricing.

AI-native cloud: serverless access to 200+ open models, dedicated endpoints, GPU instances, and agent sandboxes through unified APIs.

A single OpenAI-compatible API that routes to 300+ LLMs from every major provider, with per-key budgets and programmatic key provisioning.

Production AI gateway routing to 1,600+ LLMs with observability, guardrails, governance, and an MCP gateway in one API.

Run thousands of community and official AI models (image, video, audio, LLM) via one HTTP API with no ML infra to manage.

Fast RDU-hardware inference for large open models (up to 405B) with an OpenAI-compatible API and a free developer tier.

Serverless and dedicated inference for 200+ open models (Llama, DeepSeek, Qwen, FLUX) plus fine-tuning, aimed at production AI teams.

One API key and endpoint to reach hundreds of models across providers, with zero markup, BYOK, budgets, and native AI SDK integration.

Generative-media inference platform serving 600+ image/video/audio/3D models (FLUX, Kling, Sora) through one fast, queue-based API.

User-permissioned auth layer that lets agents securely log into and act inside third-party apps that lack APIs (SSO, OAuth, 2FA) without ever storing user credentials.

MCP runtime that handles OAuth and authorized tool-calling so AI agents can act on behalf of a specific user across 7,500+ agent-optimized tools.

Auth0's identity stack for GenAI: a Token Vault that stores and refreshes third-party OAuth tokens for agents, async (CIBA) human approvals, and fine-grained authorization for RAG.

Managed auth plus 1,000+ pre-authenticated toolkits that give AI agents per-user, scoped access to SaaS apps with automatic OAuth/token refresh.

Identity platform for AI agents and MCP: turn any app into an OAuth provider (Inbound Apps), vault and refresh third-party tokens, and give each agent a scoped identity.

AI-native integration platform with prebuilt connectors, configurable actions/flows and MCP so agents get authenticated, mapped access to your customers' SaaS apps.

Hosted, production MCP servers with built-in OAuth and multi-tenant auth across 600+ tools, plus the open-source Strata server for self-hosting.

Gives agents secure, authenticated access to thousands of prebuilt tools across 220+ integrations via custom MCP servers, with enterprise governance and normalized data.

Open-source integration platform providing OAuth, token refresh and credential storage for 800+ APIs, with each integration exposed to agents over MCP.

Unified integration API exposing 1,000+ prebuilt actions across 130+ connectors (plus an MCP server and triggers) so agents can act in CRMs, ticketing, email and messaging.

Agent tooling platform (OneTool) connecting agents to 100+ APIs and 25,000+ actions through a single SDK, with AuthKit for managed auth; community edition is open source.

Embedded integration API and managed auth that lets your app or agent connect accounts and run actions across 3,000+ apps on behalf of end users.

Modular auth stack for AI apps: puts an OAuth 2.1 authorization server in front of MCP servers, adds delegated agent identity and a token vault, plus enterprise SSO/SCIM.

Turns your app into an OAuth 2.1/OIDC provider giving agents and MCP servers scoped, revocable, auditable access with a drop-in consent UI and human-in-the-loop approvals.

Cloud tool infrastructure and Agent Studio that equip any LLM with hosted, low-latency tools (code exec, web, RAG, MCP) by wrapping your completion in a few lines.

With one config value AuthKit becomes an MCP-compliant OAuth 2.1 authorization server; adds SSO, SCIM, audit logs and fine-grained tool-level authorization for agents.

Open-source pricing & billing layer over Stripe — one API for entitlements, credits and usage enforcement, built for AI startups.

Subscription billing and monetization for B2B SaaS and AI companies — mix usage, credits, outcome-based and seat pricing on one billing record.

Open HTTP-native payment protocol that revives the '402 Payment Required' status code — agents and apps pay per API call in stablecoins with no account or login.

Agent payments infrastructure — give each AI agent a wallet, virtual card and cross-chain stablecoin rails under human-controlled delegation.

Merchant-of-Record payments and billing for AI-first companies — sell in 220+ countries with tax, compliance and disputes handled behind one API.

Open protocol for agent-led payments — cryptographically-signed 'mandates' let agents transact on a user's behalf across cards, bank transfers and stablecoins (via the x402 extension).

Mature open-source subscription billing and payments platform (Java) — self-host with a decade-deep plugin ecosystem for complex billing.

Open-source metering and usage-based billing API — consumption tracking, subscriptions, payment orchestration and revenue analytics, self-host or cloud.

Usage-based billing and enterprise contract management for high-volume AI/software companies (now a Stripe company).

Usage-based billing and revenue platform for high-volume AI/software — event metering, price modeling and automated invoicing.

Results/outcome-based billing for AI agents — attach value to what an agent does, track cost & margin, and invoice on outcomes.

Open-source toolkit plus a remote MCP server that exposes PayPal payments, invoices, subscriptions, orders and disputes to AI agents.

Agent-native wallet and payouts — let AI agents send and hold money under programmable, human-in-the-loop spend policies.

Open-source Merchant of Record — payments, subscriptions and usage-based billing with global tax handled for you, built for the AI era.

Entitlements, feature gating and usage-based packaging on top of Stripe — ship pricing and plan changes without shipping code.

Identity + payments rails for autonomous agents — Know-Your-Agent (KYA) identity, self-funded wallets and USDC micropayments via the open KYAPay protocol.

Pricing, packaging and entitlements runtime — the 'usage runtime' that decides what every customer, user and agent is allowed to do, with per-agent budget caps.

The default payments, billing and subscriptions API — now with an Agent Toolkit and Agentic Commerce suite so AI agents can accept and make payments.

Fully-managed serverless hosting for the developer-favorite open-source Chroma, with the same API as local Chroma and per-database scoped keys.

Serverless vector database that runs on Cloudflare's global edge network, wired directly into Workers and Workers AI for low-latency RAG.

Reactive serverless backend-as-a-service with built-in vector search alongside your app data — well suited to real-time AI apps and agents.

Fully-managed serverless vector search for multimodal AI, built on the open-source Lance columnar format — pay only for storage you use.

Vector search built into MongoDB Atlas so embeddings live next to your operational documents — free on the M0 tier, provisionable via the Atlas Admin API.

Serverless Postgres with pgvector plus a Claimable-Postgres API that spins up a live database with zero human involvement — purpose-built as a backend for apps and AI agents.

The managed serverless vector database most RAG teams reach for — autoscales to zero, with a control-plane Admin API for programmatic project and key provisioning.

Managed cloud for the popular open-source Qdrant vector engine, with a genuinely free-forever 1GB cluster (no credit card) and a cloud management API.

Managed Redis with a built-in query engine for vector, full-text, geospatial and hybrid search — real-time speed for agent memory and RAG.

Open-source Postgres backend where pgvector is included free on every plan, with a Management API and OAuth2 for programmatically creating and managing projects.

Managed Postgres tuned for scale with pgvector + pgvectorscale and hybrid vector/BM25 search — a Postgres-native RAG backend, no per-vector pricing.

Object-storage-native search engine offering vector + full-text search at roughly 10x lower cost than in-memory vector DBs, now with self-serve signup.

Serverless, HTTP-first Vector and Redis with pay-per-request pricing and a full Developer API + MCP server for provisioning databases from code or agents.

Managed cloud for Vespa, the big-data search-and-inference platform combining vector search, ML-ranked retrieval and real-time inference at massive scale.

Managed cloud for the open-source Weaviate vector database, with built-in hybrid search and vectorizer modules and a free 14-day sandbox to start instantly.

The fully-managed cloud from the makers of Milvus — serverless vector search that scales to billions of vectors, with a free tier and startup credits.

Open-source LLM/agent observability (Phoenix) plus Arize AX, the commercial platform for production monitoring and online evals.

Eval-first AI platform unifying evaluations, tracing/logging and prompt playground for shipping quality AI products.

Open-source LLM tracing, LLM-as-a-judge evals and experiment management from Comet, with a managed cloud.

Cloud AI-quality platform on top of the open-source DeepEval framework: test management, tracing and analytics.

LLM/agent observability inside Datadog's APM suite, billing only on LLM spans with built-in online/offline evals.

All-in-one evaluate/observe/improve platform for LLM and agent apps: tracing, evals, simulations, gateway and guardrails.

Enterprise AI evaluation + observability platform where offline evals become production guardrails (now part of Cisco).

OpenTelemetry-based observability and evaluation platform purpose-built for production AI agents.

LangChain's managed tracing, evaluation and prompt-engineering platform for LLM and agent apps.

Open-source-core LLM observability with tracing, evals, prompt management and playground, offered as a managed cloud.

OpenTelemetry-native, open-source tracing and metrics for LLM, vector-DB and framework calls, with a hosted cloud.

Lightweight open-source LLM observability, prompt management and analytics, tuned for chatbots and RAG.

End-to-end GenAI platform to simulate, evaluate and observe agents, with prompt versioning and online evals.

Prompt management + LLM observability platform for versioning prompts, running evals and tracing agents in production.

LLM reliability platform built on OpenLLMetry, an OpenTelemetry-native tracing layer for LLM apps.

Weights & Biases' LLM observability and evaluation toolkit that auto-traces calls and runs LLM-judge scorers.

Managed persistent memory service for agents built on Durable Objects + Vectorize, handling extraction, storage, search, and summarization.

Cloud-native context layer that ingests documents, audio, video, and web data into semantic memory retrievable by agents via one API.

Open-source SDK from LangChain that gives LangGraph agents long-term memory (extraction, search, prompt optimization) in your own store.

Fully-managed API for stateful agents that manage their own memory like an OS (context = RAM, archival = disk) and self-improve over time.

Universal memory layer that adds persistent, personalized long-term memory to AI agents and apps in a few lines of code.

User-profile-based long-term memory that batches conversations into structured profiles to personalize LLM apps cheaply and fast.

Managed knowledge/context layer on Pinecone that ingests files and serves grounded chat + context retrieval for production AI apps.

Redis-based dual-tier memory (working + long-term semantic) server and managed context engine for agents, with LangCache semantic caching.

Hosted universal memory API that stores, indexes, and retrieves long-term context, RAG, and user profiles for AI apps with one API.

Google Cloud's managed long-term memory for agents (Agent Engine) that uses Gemini to extract facts and preferences scoped per user.

Temporal knowledge-graph memory for agents that tracks what was true and when, engineered from chat history and business data.

Open-source AI memory platform that builds a knowledge-graph-plus-vector memory for agents via a self-improving cognify pipeline.

On-demand serverless sandboxes (gVisor isolation) with GPU support, persistent snapshots, and an open-source core (beta9) you can self-host as BYOC.

Perpetual-sandbox agent runtime — spin up isolated sandboxes in milliseconds (resume ~25ms with memory intact), plus hosted agents, MCP servers, and batch jobs.

GA (April 2026) container-backed sandboxes that give Workers-based agents a persistent isolated computer — exec, code interpreters, PTY terminals, backup/restore — on Cloudflare's edge.

Programmatic API over CodeSandbox's microVMs that lets agents create, run, hibernate, and instantly fork() a live VM (with memory snapshot) for parallel/A-B agent runs.

Elastic, stateful sandboxes for AI-generated code that cold-start in ~90ms, with parallel forking, snapshots, and SDKs in five languages.

A REST API that boots ephemeral Firecracker microVMs from OCI containers in a few hundred ms — a low-level sandbox layer for agents (see also Fly Sprites).

Full Linux VMs plus Git for code your AI writes — serverless Runs bill per millisecond of execution and support external npm modules and env vars for agent code.

Sandbox VMs built around Infinibranch — snapshot, branch, and restore an entire running environment (files + processes + memory) in under 250ms for parallel agent forks.

Developer platform with secure microVM sandboxes (Kata/gVisor) for multi-tenant AI code execution, self-serve BYOC across AWS/GCP/Azure and public pricing.

Enterprise-grade Devboxes for AI coding agents on a custom bare-metal hypervisor — ~100ms command exec, ~25ms resume-from-standby at zero idle cost.

Ephemeral Firecracker microVMs for running untrusted/agent-generated code on Vercel — Node & Python runtimes, root access, persistent-by-default, millisecond starts.

Open-source terminal pair-programmer that edits your local git repo with AI — not hosted; you bring your own model key.

Sourcegraph's agentic coding tool with a headless CLI and streaming-JSON output for automation — pay-as-you-go with no markup.

StackBlitz's in-browser agent that prompts, runs, edits, and deploys full-stack apps using WebContainers — no local setup.

Open-source autonomous coding agent (VS Code/JetBrains extension, CLI, and SDK) that runs on any LLM you plug in — fully BYOK.

Cloud coding agents you run at scale via a Python SDK/REST API, wired into GitHub, Slack, and Linear to open PRs autonomously.

AI code editor whose agent is now scriptable via a headless CLI, TypeScript SDK, and cloud Background Agents for developers and CI.

Autonomous software-engineer agent that runs long-horizon tasks in its own cloud VM and is triggerable via a REST API from CI, hooks, or cron.

Agent-native dev platform whose Droids run in managed cloud sandboxes; a headless `droid exec` CLI drops them into CI and scripts.

Google's async coding agent that works on your GitHub repo in a cloud VM, now with an alpha REST API for programmatic sessions.

AI app builder that ships full-stack products from chat, with backend/auth/DB — popular with non-technical founders.

OpenAI's coding agent across CLI, IDE, and cloud; `codex exec` runs headless and the gpt-5-codex models are callable via the OpenAI API.

Browser-based agent that builds, runs, and deploys full apps end-to-end inside Replit's cloud IDE for non-technical builders.

Agentic terminal that runs multi-agent coding/debugging workflows from natural language, with BYOK model support on every plan.

Agentic IDE (formerly Codeium) whose Cascade agent does multi-file, codebase-aware autonomous edits.

Vercel's generative UI/app agent with a real Platform API (TypeScript SDK + REST) to create chats and generate React/frontend code programmatically.

Serverless visual workflow orchestrator for AWS — coordinate Lambda, containers, and 200+ services as durable state machines with built-in retries, error handling, and history.

Serverless durable-execution engine on Cloudflare Workers for long-running, multi-step apps and AI agents — steps persist across failures, sleeps, and retries at the edge.

Postgres-backed durable execution as a library — annotate ordinary Python/TS/Go/Java code with decorators to get exactly-once workflows, queues, and recovery, hosted serverlessly on DBOS Cloud.

Managed, asset-based orchestration for data, ML, and AI pipelines — declarative software-defined assets with lineage, scheduling, and serverless or hybrid compute.

Postgres-backed orchestration engine for background tasks, AI agents, and durable workflows — a high-throughput task queue with retries, DAGs, and real-time monitoring.

Event-driven durable functions and AI workflows for TypeScript/Python/Go — steps, retries, sleeps, and flow control with zero queues to run.

Managed hosting for Kestra, the declarative YAML-based orchestration platform — event-driven data and infra workflows with 600+ plugins and a code-friendly UI.

Managed deployment and orchestration for stateful LangGraph agents — persistence, human-in-the-loop, streaming, and horizontal scaling for production AI agents.

One-command managed deployment for Mastra, the TypeScript AI-agent framework — agents, tools, workflows, and memory with routing, scaling, and tracing handled for you.

Managed, enterprise Netflix Conductor — a distributed orchestration platform for microservices, workflows, and AI agents with a visual builder and multi-language SDKs.

Workflow automation plus Connect — SDKs/APIs that let your app or AI agent add 2,500+ integrations and provision end-user account connections and OAuth tokens programmatically.

Python-native workflow orchestration for data, ML, and agents — turn functions into observable, retryable flows with scheduling, and a managed control plane plus MCP gateway.

Durable-execution runtime for resilient services, workflows, and AI agents — durable RPC, virtual objects, and state, deployable on serverless (Cloudflare/Vercel/Deno) or as managed Restate Cloud.

Fully managed durable-execution platform where you write Workflows as ordinary code that survive crashes, retries, and multi-day sleeps.

Open-source TypeScript platform for long-running background tasks and AI agents with retries, queues, concurrency, and observability, fully managed in the cloud.

Open-source developer platform that turns scripts (TS/Python/Go/Bash/SQL) into workflows, webhooks, and internal UIs — a fast workflow engine and Retool/Temporal alternative.

Managed hosting for n8n, the fair-code visual workflow automation and AI-agent builder with 400+ integrations, native AI/LangChain nodes, and an MCP server.

IBM's MIT-licensed document parser — runs DocLayNet layout and TableFormer table models locally on commodity hardware, no cloud egress.

Open-source library that turns 25+ file types into semantically labeled elements (title, table, list) with positions — the preprocessing layer for RAG ingestion.

LlamaIndex's managed document parser — per-page tiers from fast heuristics to VLM-agentic, with native LlamaIndex ingestion for RAG.

Agentic document parsing — layout-aware vision + VLMs + a multi-pass correction loop turn messy PDFs, scans, and spreadsheets into structured, RAG-ready data.