Brex Uses Claude Code to Build Text-to-SQL Tool and Integrate Company Style Guidelines.

Brex has published details on how its engineering, content design, and data teams use Claude Code to improve productivity and code quality across the organization.

According to Anthropic's case study, Brex's Product AI team now tests multiple technical approaches before committing to a solution, allowing engineers to evaluate trade-offs more quickly. The AI-assisted coding tool has also enabled Brex's content designers to systematically apply the company's style guidelines across products, a task that traditionally required manual review and editing.

The most concrete example of Claude Code's impact at Brex is Brex Explorer, a text-to-SQL interface developed by the company's Data & Analytics team. The tool allows employees to query company data using natural language rather than writing SQL manually.

Brex Explorer addresses a common bottleneck in data-driven organizations: analysts and non-technical employees often need to wait for database queries to be written by engineers. By translating plain English requests into SQL, the tool reduces time to insight and makes data more accessible across teams.

The case study does not specify the technical architecture of Brex Explorer or whether it uses Claude's API directly, but the timing aligns with Anthropic's recent push into enterprise AI tooling. Claude recently claimed that its tools could save users 80% of time on certain tasks, and Brex's implementation offers a practical validation of those productivity gains in a real-world setting.

Beyond engineering workflows, Brex's content design team uses Claude Code to maintain consistency across the company's products. Style guides are notoriously difficult to enforce at scale, particularly in fast-moving startups where multiple teams ship features independently.

By integrating guidelines directly into the development process through Claude Code, Brex reduces the need for post-hoc editing and ensures that user-facing copy adheres to brand standards from the start. This approach mirrors how some engineering teams use linters and automated checks to enforce code quality, but applies the same rigor to language and tone.

Brex's adoption of Claude Code reflects broader trends in how companies are integrating AI into their workflows. Rather than treating AI as a standalone product, organizations like Brex are embedding it into existing tools and processes, from code review to data analysis to content production.

Anthropic has positioned Claude as an enterprise-grade alternative to other large language models, emphasizing safety, reliability, and integration with developer workflows. The Brex case study provides concrete evidence of how those features translate into operational improvements, particularly for teams that need to balance speed with quality.

The company has not disclosed whether other teams at Brex use Claude Code or what specific metrics they track to measure its impact. However, the examples shared suggest that AI-assisted coding is moving beyond novelty use cases and into the core workflows of high-performing engineering organizations.