Use cases
What teams build
Examples of agents teams stand up with Chat.dev. Each is something you build and configure, working through the integrations and tasks you set up.
Coding
An agent that lives in the IDE and answers from Slack, reads a repo over SSH, runs builds on its own machine, and holds context across a project.
GTM
An agent that researches accounts, drafts outreach, and prepares materials, using the tools and data you connect to it.
Retention
An agent you set up to watch for the signals you define and follow up through the channels you connect.
Data science
An agent with a persistent workspace for notebooks and datasets that can run analyses and share results with the team.
A production incident from your phone
An example from the product tour: checkout starts returning errors while you are on a train. Call Werner, point an agent at the repository and have it investigate a migration, prepare a patch, run tests and report the result. Review the change and customer follow-up from your phone.
Build and review an API
Give an agent a task such as building a REST API with authentication, rate limiting and tests. Follow its terminal output and files as it works, then review its diff and send line comments back as the next task.
Prod went down while you were on the train.
An illustrative scenario from the product tour.
You're stuck.
It's 8:14. Checkout is throwing 500s, customers are bouncing, and all you have is a phone with one bar of signal. Nothing you can do the old way.
Incident · checkout · 8:14 AM
POST /api/checkout → 500 (92% error rate) stripe webhook handler: unhandled exception customers affected, last 10 min: 38
You call Werner.
You dial +1-856-CHATDEV and tell him what's happening. He spins up a Claude Code agent on your own Anthropic subscription, points it at the repo, and stays on the line with you.
You: Prod's down, checkout is 500ing.
Werner: Spinning up Claude Code on your Anthropic plan and pointing it at the repo now.
You get to the office to good news.
The agent traced the bad migration, shipped the patch, watched the deploy go green, and emailed the 38 customers who hit the error to say it's resolved. Before your train even arrived.
$ revert + patch migration 0042 deploy: succeeded in 47s POST /api/checkout → 200 (error rate 0%)
Emailed 38 affected customers: it's fixed