# How Vouch uses Greptile to enhance code quality in their monorepo

|  |  |
| --- | --- |
| 85% | Reduction in review time |
| 2,000+ | Critical issues caught |

See how Greptile can enhance your code quality

[Request a demo](/contact)

|  |  |
| --- | --- |
| Dan Goslen | Senior Software Engineer III |
| 200+ | Company Size |
| Series D | Series |
| InsurTech | Industry |
| San Francisco, CA | Headquarters |

> “Engineers have mentioned having more confidence in their ability to implement a new feature or bug fix by utilizing Greptile's understanding.”

Senior Software Engineer III @ Vouch

Dan Goslen

## Q&A

### What specific technical challenges did Vouch face?

With multiple internal platforms and APIs to maintain in a complex monorepo, Vouch faced challenges with PR review lead times and catching subtle implementation errors that were difficult for human reviewers to detect consistently.

### Why did Vouch choose Greptile over other AI tools?

Vouch chose Greptile for its minimal setup requirements and immediate value delivery. The team particularly appreciated the ability to tune prompts for their monorepo structure and the flexibility to control usage through labels.

### What types of issues has Greptile caught for Vouch?

Greptile has helped catch critical issues around incorrect typing, poor asynchronous handling, improper logging patterns, and incorrect API usage that static analysis tools would not be able to detect.

### How has Greptile improved developer experience at Vouch?

The integration has enabled developers to receive immediate feedback on code issues, allowing human reviewers to focus on architectural concerns and best practices. The chat feature has become a valuable tool for understanding complex domain aspects of the system.

\[ TECHNICAL CHALLENGE \]

## Maintaining code quality in complex monorepo architecture

How Vouch leverages AI to catch subtle implementation errors and improve review efficiency

With multiple internal platforms and APIs to maintain, Vouch's engineering team faced challenges with PR review lead times and catching subtle implementation errors that were difficult for human reviewers to detect consistently. “We saw that some of our most common mistakes were from issues that human reviewers have a hard time reliably detecting but are easy for a trained automation tool to catch every time,“ says Dan Goslen.

After evaluating several AI tools, Vouch chose Greptile for its minimal setup requirements and immediate value delivery. The team particularly appreciated the ability to tune prompts for their monorepo structure and the flexibility to control usage through labels. “We've tuned our prompt to make sure our engineers are using these packages properly in addition to looking for style and logic issues.“

Greptile has helped Vouch catch critical issues around incorrect typing, poor asynchronous handling, and improper logging patterns. The code review summary feature has enhanced human reviewer effectiveness by providing clear context before deep-diving into the code. “We get comments like 'This package exposes an API that looks like this, but you are using it incorrectly', which is very useful. Static analysis tools would not be able to help in that situation.“

\[ KEEP READING \]

## Learn more about Greptile

[See docs](https://www.greptile.com/docs/introduction)

### [AI Code Reviewer](/agent.md)

The code reviewer that catches real bugs

[Learn more](/agent.md)

### [TREX: Runtime Validation](/trex.md)

Test every code change

[Learn more](/trex.md)

### [Independence](/independence.md)

The independent code validator

[Learn more](/independence.md)

### [Learning](/learning.md)

Code reviews that get personalized over time

[Learn more](/learning.md)

## Greptile consistently catches more bugs in large repos

[Contact Sales](/contact)

[Start now](https://app.greptile.com/signup)