Autonomous AI software engineer that plans work, writes and tests code, and opens pull requests for review.
Sold directly by the vendorConnect with vendor →
Autonomous AI coding agents that work on tasks for you
Coding agents take a task such as fixing a bug, plan the change and return it for review. They work best on well-defined tasks with good tests, and every change still needs a human reviewer.
SARVO lists 12 tools for this task. Tools with no implementation partner are sold directly by the vendor, and SARVO can connect you with them. Ask SARVO to match your exact need.
Tools to consider
#2 · by Anthropic
Claude Code
1 implementation partnerView partners →
#3 · by Factory
Factory Droids
AI agents for enterprise engineering teams that handle coding, review, testing and incident work.
Sold directly by the vendorConnect with vendor →
#4 · by All Hands AI
OpenHands
Open-source platform for autonomous coding agents that plan and apply changes across a project.
Sold directly by the vendorConnect with vendor →
Google's asynchronous coding agent that fixes bugs and writes features in a cloud sandbox.
Sold directly by the vendorConnect with vendor →
Open-source coding agent for VS Code that edits files and runs commands with your approval.
Sold directly by the vendorConnect with vendor →
Open-source command-line pair programmer that edits code in your git repository.
Sold directly by the vendorConnect with vendor →
#8 · by GitHub
GitHub Copilot
Sold directly by the vendorConnect with vendor →
#9 · by AWS
Amazon Q Developer
AWS coding assistant and agent that writes, tests, upgrades and explains code.
Sold directly by the vendorConnect with vendor →
#10 · by Augment Code
Augment Code
AI coding agent built for large codebases, with deep repository context.
Sold directly by the vendorConnect with vendor →
AI code editor whose Cascade agent follows your changes across a repository.
Sold directly by the vendorConnect with vendor →
Open-source local AI agent that installs, runs and tests code and connects to tools through MCP extensions.
Sold directly by the vendorConnect with vendor →
What to check before you choose
- Does it run in a sandbox, and what can it access in your repositories?
- Are your code and prompts used for model training?
- How good are your tests? Agents depend on them to check their own work.
- How are changes delivered: pull requests you can review and revert?
SARVO does not certify tools. Verify security and data terms with each vendor.
Compare tools
Find a partner by city
Guides
More AI tasks
Voice to textText to speechVoice agentsCoding assistantsLegalContract reviewMeeting notesDictationCustomer supportIT and HR helpdeskSales call analysisClinical notesEnterprise searchCompany AI assistantsAgent platformsGPU cloudVector databasesWriting assistantsAI sales agentsContact centre AIVoice AI platformsWorkflow automation agentsAI agent frameworksBrowser and computer agentsAI app buildersHR and recruiting AIFinance AISecurity operations AIData analytics AIAI research assistantsWorkplace AI assistants
Frequently asked questions
What is a coding agent?
A tool that plans and makes code changes across a project, runs commands and tests, and returns the result for review.
Devin or Claude Code?
Devin works as an autonomous engineer in its own environment, while Claude Code runs in your terminal and editor with you watching. Try both on a real ticket.
Are open-source agents a good choice?
OpenHands, Cline, Aider and Goose are open source, so you can run them yourself and choose your model.
Will agents replace developers?
They speed up routine work. People still decide what to build and review what ships.