Navan cut a month-long Spring Boot migration to a week with Cursor
Navan, the business travel and expense management company, ran its latest Spring Boot upgrade with Cursor's Cloud Agents: a senior engineer planned the migration once and handed each component to an agent, so 8 applications and 15 libraries moved in parallel. Cursor reports the migration finished in about a week without freezing development, where a comparable one two years earlier took over a month.
The problem
Two years ago, moving Navan's codebase to a new Spring version took one senior engineer over a month. Incomplete pushes and reverts locked the master branch for a day, other engineers were still catching up days later, and the bigger cost was a month of an engineer's time spent on the migration instead of features or bug fixes. Navan's previous coding agent also needed someone at one desktop with the session open the whole time, which ruled out parallel, asynchronous work.
How they did it
- Plan the upgrade (here to Spring Boot 4) up front and split it into components, one per application or library.
- Delegate each component to its own Cursor Cloud Agent so the applications and libraries move at the same time instead of one after another.
- Run the agents where they can reach your internal systems; Navan runs GitHub work on Cursor-managed infrastructure and internal MCPs for Snowflake and Slack on its own self-hosted machines.
- Let agents spin up a browser on those machines to click through flows and verify their own changes.
- Add Automations, Skills and Memories on top so eval loops and migrations keep running asynchronously without someone watching them.
- Keep people in charge of review; generated code goes through the same human and AI review, testing and staging as any other change, and complex changes are verified in staging before production.
Results
- A Spring Boot migration across 8 applications and 15 libraries finished in about a week, against over a month for a comparable migration two years earlier, as reported by Cursor.
- Development did not have to be frozen during the migration, as reported by Cursor.
- About 200 engineers now actively use Cursor's Cloud Agents, a 10x increase since April 2026, as reported by Cursor.
- 36% of merged core-repo pull requests over a 30-day period were authored with Cursor, as reported by Cursor.
- On the product side, one engineer took over feature work that previously needed 2 or 3 engineers across multiple sprints, as reported by Cursor.
- Investigating issues in two unfamiliar codebases saved roughly two days of ramp-up time, as reported by Cursor.
As reported by the source (Cursor customer story); AgentGid did not measure these figures.
This fits an engineering organisation with many services on a shared framework, a solid test and staging setup, and someone senior enough to plan the migration and review what the agents produce. It suits a small team with one app less: the parallelism is the point. Per AgentGid's Cursor pricing page (checked Oct 14, 2026), Cursor has a free tier with limited AI usage, individual plans from $8 a month, Teams from $30 per user a month with pooled usage and custom Enterprise pricing; running agents on your own machines also means operating those workers yourself, and Cursor's docs recommend its managed Cloud Agents for most teams. All figures come from Cursor's own customer story and are not measured independently; the month-long baseline is a different migration from two years earlier, done by one engineer with older tools. If you want cloud coding agents without a Cursor seat, OpenAI Codex is included in ChatGPT plans from $8 a month per its AgentGid page (checked Oct 6, 2026), and its CLI is open source.
The agent used here
Similar use cases
Cursor's planner-worker agents built a web browser from scratch in about a week
Cursor tested long-running autonomous coding with a hierarchy of planner agents that create tasks and worker agents that just complete them. The headline project was a browser eng…
The agents ran for close to a week, writing over 1 million lines of code across 1,000 files.
How PlanetScale uses Bugbot as an automated first pass on code review
PlanetScale added Cursor's Bugbot to its pull request workflow to catch logic issues before merge. As reported by Cursor, engineers act on most of its comments.
As reported by Cursor: 'roughly 80% of Bugbot comments are addressed before merge time'.
How National Australia Bank sped up legacy migrations with Cursor
NAB engineers used Cursor to understand and migrate legacy Silverlight/.NET and Assembly systems, and to build a new payment app. As reported by Cursor, delivery was well ahead of…
As reported by Cursor: NAB 'originally scoped six months of work' for the BizCalc migration and now expects it to finish in two months, 'a 3x improve…
Guides
Malicious Claude Code 2.1.313 Extension in Cursor: What It Did and How to Check, Checked
A tampered Claude Code extension, version 2.1.313, sat in Cursor's extension marketplace on October 29, 2026 and tried to plant an SSH backdoor. What Cursor's notice confirms, why the tunnel mostly failed, and how to check and clean a machine.
How to Cut Token Usage in Claude Code and Cursor (2026)
Practical ways to spend fewer tokens with Claude Code and Cursor: measure first, keep context small, match the model to the task, filter what the agent reads, and the open-source tools that help.
VoiceStudio: The Free, Local ElevenLabs Alternative, Checked
A viral post calls VoiceStudio a 100% free ElevenLabs replacement with 19.4K stars. We checked it against GitHub and the project's own pages: what it does, what is really free, the license catch in the default voice model, and how to plug it into Claude Code or Codex.
AI Coding Agent Pricing Explained: Claude Code, Cursor, Codex and Copilot (2026)
What Claude Code, Cursor, OpenAI Codex, GitHub Copilot and other coding agents really cost: plan prices, usage limits, subscription vs API billing, and how to estimate your monthly bill.