Vertical Autonomous Applications
Production agents shipping code and running operations across business functions.
Coding & Software Engineering Agents
23AI coding platform specializing in code integrity — test generation, code review, and coverage analysis. Agents analyze pull requests, suggest tests, and identify edge cases automatically. Focused on correctness rather than just code generation speed.
Open-source platform for AI software agents (formerly OpenDevin). Agents can write code, run commands, browse the web, and call APIs to complete full software engineering tasks. Ships with a web UI and supports 10+ agent strategies.
Anthropic's API for letting Claude control a computer: take screenshots, click, type, and run terminal commands. The first production-grade computer-use API from a frontier lab. Enables building agents that can operate any desktop software.
Terminal-based AI coding engine for large tasks spanning many files. Uses a protected sandbox for making changes before applying them. Designed for complex, multi-step coding projects where other agents lose context.
GitHub's AI tool for building micro-apps from natural language descriptions. Generates functional web apps without requiring any code, hosted on GitHub. Part of GitHub Next's exploration into AI-first development workflows.
Together AI's open-source code generation app powered by Llama 3.1 405B. Generates complete React applications from a single text prompt. Demonstrates the capability of open-weights models for full-stack code generation.
AI coding assistant that understands your entire repository and makes multi-file changes from the command line. Unlike tools that work file-by-file, Mentat considers the full codebase context. Useful for large-scale refactoring and cross-cutting feature additions.
Amazon's AI coding assistant with deep AWS integration for building, testing, and deploying cloud applications. Understands AWS services and best practices natively. Includes a transformation agent for automated Java upgrades and security vulnerability patching.
AI software engineering platform deploying Droids — specialized agents for code review, refactoring, security scanning, and documentation. Integrates with GitHub to act as an automated team member on pull requests. Built for enterprise engineering teams.
AI full-stack engineer that builds production-ready web applications from a description. Generates React frontends with Supabase backends and deploys to the web instantly. Iterated through conversational edits with live preview — popular with non-developers.
Open-source system for generating and iteratively executing code, inspired by GPT-4 Code Interpreter. Fine-tuned models that excel at data analysis, visualization, and math. Provides a self-hosted alternative to commercial code interpreters.
Open-source AI code assistant for VS Code and JetBrains with full agent mode. Supports any LLM (local or cloud) and lets you define custom slash commands and context providers. Community-driven alternative to Cursor with complete model flexibility.
Self-hosted AI coding assistant with code completion, chat, and agent capabilities. Runs on your own hardware or private cloud with consumer GPU support. Provides the core features of GitHub Copilot without sending code to third parties.
Free AI code acceleration toolkit with code completion and search across 70+ languages and 40+ IDEs. Uses its own models to provide fast, accurate suggestions. Windsurf IDE from Codeium ships a full agent mode with multi-file editing.
Agent-powered dev environment that goes from issue to pull request autonomously.
Microsoft's AI coding assistant with agent extensions. Copilot Workspace handles issue-to-PR workflows autonomously. Copilot Extensions allow third-party tools to be integrated as @-mentioned agents in the chat. Deep GitHub integration with repository context.
Open-source autonomous software engineering agent from Princeton. Uses a custom Agent-Computer Interface (ACI) optimized for LLM code editing tasks. Set state-of-the-art performance on SWE-bench — the standard benchmark for autonomous bug fixing in real GitHub repos.
VS Code extension giving Claude (and other LLMs) access to your full development environment. Reads files, writes code, runs terminal commands, and browses the web with human approval gates. Open-source with full transparency into what the agent is doing.
Open-source AI agent from Block (formerly Square) that lives in your development environment. Automates engineering tasks across files, shells, and browsers. Plugin architecture (extensions) for custom tools. Designed to work autonomously on multi-step tasks with minimal hand-holding.
AI IDE from Codeium with a "Cascade" agentic system for multi-file, multi-step code changes. Tracks codebase changes in real-time and maintains context without manual pinning. Integrated search, terminal, and browser. Competitive pricing compared to Cursor.
Terminal-first AI coding agent from Sourcegraph. Reads your entire codebase context, plans multi-step changes, and executes them in the terminal. Open-source under Apache 2. Strong support for large monorepos and enterprise codebases with sophisticated context retrieval.
Anthropic's official CLI agent for software development tasks. Runs in your terminal, reads your codebase, and implements changes based on natural language instructions. Integrates with Claude's full context window for understanding large codebases.
Google's enterprise AI coding assistant with agentic capabilities. Integrated into VS Code, IntelliJ, and Cloud Shell. Agent mode for autonomous multi-file changes. Deep Google Cloud and BigQuery integration. Enterprise-grade with data privacy guarantees and audit logs.
AI Software Engineers (DevOps & Coding)
16The first commercial AI software engineer by Cognition. Plans, codes, debugs, and deploys complete software projects autonomously. Operates in its own sandboxed computer with browser, terminal, and IDE access. Designed for real-world engineering tasks — not just autocomplete.
GitHub Next project that converts issues directly into pull requests. Analyzes the issue, plans the required changes, implements them across multiple files, and opens a PR — all from one click in the GitHub UI. Deep integration with the repo context and CI/CD pipeline.
AI-first code editor with a powerful agent mode that reads and edits entire codebases from natural language descriptions. Multi-file editing, terminal commands, and web search integration. Built on VS Code — familiar UX with a step-change in AI capability.
AI coding assistant with deep codebase context from Sourcegraph's universal code search. Understands large monorepos that exceed LLM context windows by fetching relevant code snippets on demand. Supports Claude, GPT-4, and Gemini backends with IDE plugins for all major editors.
AI developer agent that turns GitHub issues and feature requests into pull requests automatically. Reads your codebase, understands the context, makes targeted changes, writes tests, and responds to review comments — reducing the time from issue to merge.
AI code completion and generation tool with enterprise-grade privacy. Self-hosted deployment option keeps code off third-party servers. Trained on permissively licensed code. Integrates with all major IDEs. Strong enterprise adoption for teams with IP or compliance concerns.
Open-source autopilot IDE extension for VS Code and JetBrains. Use any model (Claude, GPT-4, local Ollama) for code completions, chat, and edit actions. Highly customizable with context providers, slash commands, and model providers. The open alternative to Copilot with full model flexibility.
Enterprise autonomous software engineering platform powered by specialized Droids. Factory agents handle code review, bug fixes, dependency upgrades, and documentation — integrating with existing CI/CD pipelines to automate the recurring maintenance work that slows engineering teams.
Open-source AI pair programming tool for the terminal. Edit code across multiple files with git integration — changes are automatically committed. Supports all major LLMs and handles large codebases with intelligent context management. Actively maintained with weekly releases.
Research lab behind Devin and the team pushing the frontier of autonomous coding agents. Focused on building agents that can reason about codebases at human-expert level — their research advances are regularly published and heavily influence the broader coding agent ecosystem.
GPU-accelerated collaborative code editor with built-in AI capabilities. Inline edit, AI panel chat, and context-aware completions. Written in Rust for exceptional performance. Supports Claude and custom language server protocols. Pair programming mode lets two developers share a session with AI assistance.
AI-powered code quality analytics platform for engineering organizations. Analyzes commit history, PR velocity, churn rates, and code complexity to surface actionable insights — helping engineering managers understand the health of AI-assisted codebases over time.
StackBlitz's AI-powered full-stack web development environment in the browser. Describe an app and Bolt builds, runs, and deploys it instantly. WebContainers run Node.js natively in the browser without a backend. Full npm support and one-click Netlify deployment.
AI full-stack engineer that builds complete web applications from a chat interface. Generates React frontends with Supabase backends, deploys to production, and iterates based on feedback. Handles authentication, databases, and styling — the fastest path from idea to shipped product.
Full-featured AI coding assistant integrated into Replit's cloud development environment. Complete AI agent, code completion, debugging assistance, and deployment in one platform. Specializes in getting beginners to their first working application with minimal friction.
Enterprise & Operations (SDR / HR / Marketing)
35Web-based AI agent platform for research, data enrichment, and multi-step task automation. Agents browse the web, process PDFs, and output structured results to spreadsheets or databases. Designed for business analysts and operations teams.
Platform for deploying AI agents that automate repetitive business processes end-to-end. Agents integrate with existing enterprise tools (Salesforce, Zendesk, SAP) and handle exceptions intelligently. Includes a visual workflow builder and ROI dashboard.
AI layer for databases enabling agents to query structured data with natural language and train ML models directly in SQL. Connects to 200+ data sources and exposes them as virtual AI tables. Bridges SQL workflows with modern LLM capabilities.
Building autonomous AI employees starting with Ava, an AI sales development representative. Ava researches prospects, writes personalized outreach emails, manages follow-up sequences, and books meetings — running entire outbound sales workflows without human intervention.
Deployless backend platform where code runs immediately without builds or servers. Increasingly used for hosting lightweight AI agent backends and webhooks. Instant deploys and trace-based debugging make it ideal for rapid agent prototyping.
No-code security automation and AI agent platform trusted by 600+ security teams. Automates alert triage, incident response, and threat intelligence workflows. Recently added LLM-powered natural language building and AI analyst agents.
AI data extraction platform for automating document processing pipelines. Extracts structured data from invoices, contracts, and forms with high accuracy. Agents validate extracted data and route exceptions to human reviewers.
AI agent workspace embedded into a collaborative project management tool. Build agents that can research topics, generate content, manage projects, and interact with tasks. Agents live alongside human teammates in shared workspaces.
Enterprise agent platform for building and deploying AI agents within internal tools and workflows. Provides pre-built agent templates for HR, finance, and sales. Offers on-prem deployment for organizations with strict data residency requirements.
AI agents for enterprise knowledge work — taking actions in web apps and desktop software on behalf of users. Built on a custom multimodal model trained on software interaction. Deployed in partnerships with major enterprise software vendors.
Developer platform for building, testing, and deploying voice AI agents. Provides a WebSocket API and dashboard for orchestrating real-time voice conversations. Supports custom LLMs, STT models, and TTS voices for fully custom voice agent stacks.
Platform for building AI voice agents that handle phone calls with human-like conversation. Sub-1-second response latency with sentiment detection and call transcripts. Handles inbound/outbound calls for scheduling, customer service, and surveys.
Collaborative data workbook platform with AI agents for writing SQL, Python, and creating visualizations from natural language. Agents understand the full notebook context and connected data warehouse schema. Built for data teams doing exploratory analysis.
Browser extension and workflow automation tool with AI agents that can scrape web data, interact with SaaS apps, and automate repetitive browser tasks. 4M+ users use it to automate sales research, outreach, and data entry workflows.
AI chief of staff that manages calendars, meeting notes, and action items across an organization. Listens to meetings, captures decisions, and follows up on commitments automatically. Designed for executives who need AI-powered operational leverage.
No-code platform for building AI-powered digital workers. Lindy agents can schedule meetings, draft and send emails, handle customer support, process invoices, and coordinate across tools — a general-purpose AI employee builder for non-technical teams.
Zapier's AI layer enabling natural language automation creation and AI-powered Zap steps. Describe a workflow in plain English and Zapier builds it. AI actions let agents trigger 6,000+ app integrations from conversational interfaces. No-code agent building for business users.
No-code AI employee platform for business automation. Create AI assistants that manage email, schedule meetings, update CRM records, and handle customer support. Connects to 3,000+ apps. Lindy agents can spawn sub-agents to delegate tasks for complex workflows.
Autonomous AI agent for business research and analysis tasks. Conducts deep web research, synthesizes information from multiple sources, and produces structured reports. Task-based interface lets business users delegate research projects without understanding prompt engineering.
No-code platform for building and deploying AI agents and multi-agent teams for business use. Visual agent builder with pre-built templates for sales, support, research, and operations. Tool library for connecting agents to databases, APIs, and knowledge bases.
Custom AI agents for internal company operations connected to your data sources. Build agents for HR, engineering, sales, and support that know your company's context. Handles Notion, Confluence, Slack, GitHub, and more as knowledge sources with granular access controls.
Autonomous data labeling agent framework from HumanSignal (Label Studio). Agents learn labeling skills from human feedback and apply them at scale. Iteratively improves accuracy by checking work and requesting guidance when uncertain. Dramatically reduces manual annotation costs.
Platform for building and deploying AI agents connected to business data and actions. Visual workflow builder with JavaScript code steps for custom logic. Pre-built integrations for customer support, sales, and HR workflows. Handles auth, memory, and tool orchestration.
Open-source framework for building and running AI assistants and agents in production. Managed service or self-hostable. Supports RAG, tool use, memory, and multi-agent workflows. REST API and SDKs for Python and TypeScript. Good starting point for custom vertical AI agents.
AI-powered data enrichment and outreach platform used by sales and growth teams. Agents enrich prospect data from 75+ data sources, write personalized outreach based on enriched context, and automate follow-up sequences. A concrete example of vertical AI agents driving measurable business ROI.
AI-powered workspace for business teams. Automate repetitive tasks across email, CRM, and productivity tools with AI agents. Knowledge base integration ensures agents answer with company-specific information. Workflows combine multiple AI agents for complex business processes.
Predictive AI platform that combines machine learning with database-style query capabilities. Agents can use Aito to make probabilistic decisions — predicting the best next action, categorizing inputs, or recommending options — without training custom ML models from scratch.
Visual AI automation builder where you drag-and-drop AI nodes into pipelines. Web scraping, document parsing, LLM processing, and API integrations in one visual canvas. Designed for non-engineers to build powerful AI workflows without code.
AI-native IDE for building AI applications and agents collaboratively. Combines the ease of a doc editor with the power of a prompt engineering environment. Teams can iterate on agent prompts together, share results, and deploy to API endpoints with versioning.
Open-source browser automation agent using LLMs and computer vision to navigate websites. Doesn't require hardcoded XPath selectors — understands pages visually and semantically. Handles dynamic pages, CAPTCHAs, and multi-step workflows that break traditional scrapers.
Web-based autonomous AI agent platform where you set a goal and the agent plans, executes, and iterates to accomplish it. Open-source codebase with a polished UI. Supports multiple LLM backends and tools. Popular starting point for understanding autonomous agent behavior.
AI agent platform for automating web tasks through natural language instructions. Handles login, form filling, data extraction, and multi-step workflows across any website. API available for integrating web automation into existing agent pipelines.
Open-source conversational AI platform for building enterprise chatbots and agents. Visual flow builder with LLM integration, knowledge base Q&A, and 100+ channel integrations (Web, Teams, Slack, WhatsApp). Strong NLU capabilities and extensive analytics dashboard.