The ticket is the interface: how our developers ship without a laptop
Neither I nor my developers need a laptop to ship code anymore. That sounds like a stunt, so here is the boring version: on our team a developer's working day happens inside a ticket thread, and the parts that make that possible are mostly scripts, not AI. This post is the long version of what I showed in a short video on LinkedIn. It covers what the pipeline does, how the pieces fit, what we…
Our software team at Oizom, an environmental monitoring company, has transformed the developer experience by eliminating the need for laptops and leveraging AI-powered coding agents within ticket threads. This approach allows developers to ship code without ever opening an IDE or terminal.
The team consists of various repositories, including hardware drivers, a simulator, a backend, a frontend, and more. Traditionally, getting a working local setup took new developers days, and even experienced developers struggled with the ubiquitous question: "Does it work on your machine?"
The AI coding agents proved to be fast, but the bottleneck was the waiting time for the agent to complete the task. The developers aimed to achieve five goals: enabling developers to hand off tasks and work on other tasks simultaneously; ensuring each task runs in isolation to avoid conflicts and allow features to progress on different timelines; maintaining a linear engineering process, including branches, commits, pull requests, reviews, and CI; storing every prompt with the work it produced for review and coaching; and using the ticket as the audit trail for every aspect of the development process.
The company already had a ticket tracker, so they decided to use tickets as the only interface for developers. A developer writes their specifications in plain English and moves the ticket to "Approved" status. Each developer's workspace features their own devbox: VS Code running in the browser, with an orchestrator inside. Orchestrator assigns each ticket to an AI coding agent and ensures the agent only works on tasks assigned to that developer.
When an approved ticket is received, Orchestrator provisions a workspace for it, including a fresh branch in every repository, a copy of the database, and a private preview URL with a self-generated HTTPS certificate. This setup allows developers to work on multiple tickets simultaneously without interference from one another. If an agent causes any issues, it only affects the copy of the data, not the actual production environment.
The AI coding agent picks up the ticket, receives the ticket text and the workspace, and proceeds with the engineering work. The developer only interacts with the agent through the ticket thread. Once the agent is done, it posts a preview link and moves the ticket to QA/QC. The developer reviews the preview, provides feedback in the same thread, and moves the ticket back for further work. The same agent continues working on the ticket, maintaining context from the previous interaction.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
