Automate large service upgrades with self-checking agents
Brex used Cursor agents to let one engineer update 180 microservices to a newer Micronaut release in roughly a month.
Agent Uses · 1,321 real examples
Real examples of people putting Cursor to work: the task, how it went, what to watch out for, and a prompt you can adapt.
The day's AI news plus 3 agent uses you can try, with ready-to-paste prompts. Free; unsubscribe anytime.
Brex used Cursor agents to let one engineer update 180 microservices to a newer Micronaut release in roughly a month.
A DevOps engineer used Cursor to read production pod logs through Kubernetes MCP and to repair a broken Docker Compose setup.
Kent C. Dodds used Cursor to organize site dependency work, then carried out a Remix v2 to React Router v7 migration and related updates.
A developer used Cursor with tfmcp to plan and apply Postgres Terraform changes, then fell back to the terminal when permissions blocked the first attempt.
Use a planner-worker loop with periodic review to keep coding agents moving on a large repository task.
A writer used Cursor to help draft and revise papers in Markdown while pulling citations from Zotero and a BibTeX library.
They had Cursor pull passages from Obsidian highlights and notes, then weave those quotes into article drafts.
Use an agent to write personalized outbound, follow-up emails, and a deal forecast from live sales data.
Use an agent to rename messy invoice files, file them into folders, and prepare monthly invoice emails from a PDF template and recipient list.
Cursor lets support teams gather logs, compare old cases, and draft a bug report plus customer response for review.
A Rippling engineer uses Cursor to collect scattered work items every two hours and turn them into one dashboard.
They used Cursor for personal tasks while keeping private notes and stricter rules separate from public drafting.
They used Devin to make a Gmail assistant that shows recent emails that still need replies.
A founder used Cursor to inspect a trip-planning app, surface missing pieces, sketch screens, and then build a feature with parallel agents.
They set up a local layer that makes coding agents announce planned changes so overlapping architecture work gets caught early.