What changed

Fieldwork Code launched on Product Hunt as a mobile-accessible interface for AI coding, according to its Product Hunt page. The listing describes the product as a way to handle coding work on a laptop or in the cloud from a phone.2

How it works

On its software development page, Fieldwork describes two ways to run agents. Developers can connect repositories and run parallel coding agents on their own computers, with web search included. Alternatively, they can run sessions in a cloud sandbox that continues working when the laptop lid is closed and provides a private preview link for testing changes.1

Teams can customize how each session behaves using scripts, according to Fieldwork's documentation. Those scripts define how to set up a conversation's worktree, what commands to run and what to clean up when the session is archived. Fieldwork notes that settings files from Conductor work out of the box, and teams can sort work into custom stages named for their own workflow, such as Testing and In review.1

Fieldwork ties coding agents into its broader workspace tools, the company's page explains. Teams can store versioned specifications in a Knowledge Base where both teammates and agents can read them, and track human or agent ownership in Tasks. In a product-page example, an agent reads a size-guide spec, implements the feature, runs tests and opens a pull request while noting a remaining unit-preference question.1

For security, Fieldwork states that customer data is never used to train its own or any provider's AI models, and is encrypted in transit and at rest with org-level access controls. The company reports GDPR compliance, CASA Tier 2 certification and an operating ISO 42001 AI management system. However, its SOC 2 Type II audit and ISO 27001 certification are still underway, and the page does not list dollar prices for the Cloud tier or its standard overage rates.1