Claude Blog 采集 (2026-05-28)¶
共采集 5 篇文章
📋 文章索引¶
- How Brex improves code quality and productivity with Claude Code - Oct 30, 2025 (评分: 9.5)
- How Claude Code works in large codebases: Best practices and where to start - May 14, 2026 (评分: 9.0)
- How enterprises are building AI agents in 2026 - Dec 09, 2025 (评分: 9.0)
- How Kepler built verifiable AI for financial services with Claude - Apr 30, 2026 (评分: 9.0)
- Code w/ Claude London 2026: Rethinking how we build - May 26, 2026 (评分: 8.2)
How Brex improves code quality and productivity with Claude Code¶
来源: Claude Blog 发布日期: Oct 30, 2025 采集时间: 2026-05-28 价值评分: 9.5/10 正文字数: ~7996 字符
摘要¶
Learn how Claude Code breaks down barriers between technical and non-technical teams by making programming easier and more accessible.
正文内容¶
Learn how Claude Code breaks down barriers between technical and non-technical teams by making programming easier and more accessible.
When Andy Reed joined Brex as a content designer two years ago, he never imagined he'd be writing code and building Figma plugins. "I don't know how I would have gotten even half my work done in the past month without Claude Code ," Reed said. But that's exactly what happened when Reed embraced Claude Code , Anthropic’s agentic coding solution—and his story is just one example of how the intelligent finance platform is transforming the way their company works. To learn more, we spoke with three Brex team members about how they’re using Claude Code: Hércules Gimenes, Software Engineering Lead on the Product AI team Sumeet Marwaha, Head of Data & Analytics Andy Reed, Senior Content Designer What emerged from these conversations wasn't just a story about productivity gains, though teams report 3-4x improvements on specific tasks. It's about fundamentally reimagining how the company works together to streamline workflows, inspire creative problem solving, and drive collaboration at scale. Adopting a mindset shift with agentic coding For Hércules Gimenes and the Product AI team, Claude Code represented something more profound than a new tool. "Claude Code offers a mindset shift," he explained. "Instead of being in the driver's seat, you're the reviewer of the changes that Claude introduces. You're guiding the vision and direction." This shift has redefined what "writing code" means at Brex—a reflection of the company’s AI-first principles in action. Rather than spending time on implementation details, engineers now focus on architecture and problem-solving. The result? Better code, not just faster code. "I don't think I save time using Claude Code," Gimenes said. "I refine the problem itself much more. You can try three approaches for the same problem during the same amount of time, and after the third approach, you have a much clearer understanding and a much cleaner interface." Breaking down barriers between technical and non-technical roles While the Product AI team was discovering new ways to iterate on code quality, Reed was experiencing his own revelation. As a content designer, he found in Claude Code something unexpected: independence. Previously, updating a simple string meant filing a ticket and waiting for an engineer. Now, Reed makes PRs directly, but the real transformation came when he tackled a project that had been on the backlog indefinitely: integrating comprehensive content guidelines into every component of Brex's design system. "It would have taken me probably weeks if not months doing it manually," Reed explained. Using Claude Code, he completed the project in a few days, systematically adding content-specific guidelines pulled from internal style guides, accessibility standards and external best practices. The impact rippled across teams. Engineers and designers now have content guidelines embedded directly in their tools, eliminating the need for constant cross-referencing and reducing errors before they happen. It’s a small but powerful example of intelligent finance at work: systems running in the background, allowing people to focus on higher-value, creative work. Democratizing data access across the organization Meanwhile, Sumeet Marwaha's Data & Analytics team was solving a different but equally important challenge: making data accessible to everyone at Brex, not just SQL experts. The team built Brex Explorer, a text-to-SQL interface powered by Claude Code and MCP servers. Now, a sales manager can ask in plain text: "How many customers exist with over $10 million in their Brex account in Jacksonville, Florida and were onboarded in the last year?" The system translates the query, runs it and summarizes the results. The real multiplier effect came from their AI data engineering agent. Tasks that previously required specialized knowledge, like adding data tables with all the necessary file edits and test configurations, can now be completed by any engineer. "It's like a 4x speed increase across maybe 2x more people that are contributing," Marwaha noted. Claude Code has seen 50% adoption across the org, with Marwaha hoping to reach 100% adoption by month's end. The unexpected benefits of having an “everything tool” As teams across Brex adopted Claude Code, surprising use cases emerged. During a company hackathon, Gimenes used Claude Code's headless mode to build a sophisticated submission agent in just a few hours, something that would typically take days of development. Reed discovered he could now build tools he never imagined creating, like a Figma plugin that automatically reviews designs against Brex's standards. "It feels like I have the world's best intern," he said. But perhaps the most significant unexpected benefit has been Claude Code's role as an organizational knowledge repository. With Brex's complex monorepo structure using Kotlin and Bazel, understanding how different systems connect has always been challenging. Now, Claude Code serves as an "oracle" that can answer questions about codebase functionality instantly. "Traditionally, a design meeting would involve talking about how something should work, but nobody really understands how the sausage was made," Reed explained. "Now I've got the oracle sitting on my other monitor." Scaling Claude Code best practices across the company As Claude Code adoption spread across Brex, teams developed best practices that have become standard: Structured Context Management : Each major directory in the monorepo now has its own CLAUDE.md file containing domain-specific context. New engineers can understand Mastercard integration details or banking regulations without relying on tribal knowledge. Automated Documentation : The Product AI team implemented CI/CD checks that verify when code changes might outdate documentation, prompting updates. "Having up-to-date documentation is something you can trust—it's so powerful," Gimenes emphasized. Context-Aware Commands : Teams created custom commands that automatically load relevant context. The /submit-pr command, for instance, fetches git status, recent changes, and related PR information before executing. Start with Discovery : "Being able to find what you need is not really stored in some staff engineer's brain anymore," Marwaha observed. "It's in the codebase, accessible to everyone." The future of Claude Code at Brex Teams at Brex are already planning next-generation applications: The Data team is preparing to tackle previously untouchable projects, like analyzing trillions of credit card transaction itemization data points, work that was too tedious to justify before Claude Code made it feasible. They're also developing role-specific agents, including one to power RevOps tasks. The Product AI team envisions more sophisticated context management, where documentation updates itself automatically and context remains decentralized yet accessible. From content designers building technical tools to data analysts democratizing access, Claude Code has become what Gimenes calls an “everything tool” — useful for integrating Brex’s domain knowledge with external best practices. It also highlights a powerful duality: the intelligent finance Brex delivers to customers, helping them spend smarter and move faster, is the same philosophy that shapes how Brex teams work internally, empowering every employee to build with speed and confidence, regardless of their technical background.
Learn how financial organizations like Brex drive productivity with Claude Code.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
How Claude Code works in large codebases: Best practices and where to start¶
来源: Claude Blog 发布日期: May 14, 2026 采集时间: 2026-05-28 价值评分: 9.0/10 正文字数: ~643 字符
摘要¶
The most successful Claude Code deployments share a set of recognizable patterns across configurations, tooling, and org structure. This article is part of Claude Code at scale, a new series covering best practices for engineering organizations building with Claude Code at enterprise scale.
正文内容¶
The most successful Claude Code deployments share a set of recognizable patterns across configurations, tooling, and org structure. This article is part of Claude Code at scale , a new series covering best practices for engineering organizations building with Claude Code at enterprise scale.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
How enterprises are building AI agents in 2026¶
来源: Claude Blog 发布日期: Dec 09, 2025 采集时间: 2026-05-28 价值评分: 9.0/10 正文字数: ~4035 字符
摘要¶
New research from 500+ technical leaders reveals how enterprises are deploying AI agents in 2026—and why 80% already report measurable ROI.
正文内容¶
How are organizations deploying AI agents today—and what's in store for tomorrow? We surveyed over 500 technical leaders across industries to find out.
AI agents have moved from experimental to an essential part of an organization’s tech stack. But how are enterprises actually using them? In partnership with research firm Material, we surveyed over 500 technical leaders across industries and company sizes to understand how organizations are deploying agents today and where they see opportunity ahead. The findings show a clear pattern: organizations are shifting from simple task automation to complex, multi-step workflows that span teams and business functions. What the data shows More than half of organizations (57%) now deploy agents for multi-stage workflows, with 16% running cross-functional processes across multiple teams. In 2026, 81% plan to tackle more complex use cases, including 39% developing agents for multi-step processes and 29% deploying them for cross-functional projects. Coding leads adoption. Nearly 90% of organizations use AI to assist with development, and 86% deploy agents for production code. Organizations report time savings across the entire development lifecycle: planning and ideation (58%), code generation (59%), documentation (59%), and code review and testing (59%). But the impact extends beyond engineering. Data analysis and report generation (60%) and internal process automation (48%) rank among the highest-impact use cases. Looking ahead, 56% plan to implement agents for research and reporting over the next year. Perhaps most notably, 80% of organizations report their AI agent investments are already delivering measurable economic returns. What this looks like in practice The organizations seeing results are treating agents as a core part of their infrastrucuture, not experiments. Thomson Reuters uses Claude to power CoCounsel, their AI legal platform. Lawyers who once spent hours manually searching through documents can now access 150 years of case law and 3,000 domain experts in minutes. Cybersecurity company eSentire compressed expert threat analysis from 5 hours to 7 minutes, with AI-driven analysis aligning with their senior security experts 95% of the time. In healthcare, Doctolib rolled out Claude Code across their entire engineering team, replacing legacy testing infrastructure in hours instead of weeks and shipping features 40% faster. The retail sector is seeing similar gains. L'Oréal achieved 99.9% accuracy on conversational analytics, enabling 44,000 monthly users to query data directly instead of waiting for custom dashboards. The path forward The question for leaders in 2026 isn't whether to adopt AI agents but how to scale them strategically. The data points to three primary challenges: integration with existing systems (46%), data access and quality (42%), and change management needs (39%). Nine in 10 leaders report that agents are shifting how their teams work, with employees spending more time on strategic activities, relationship building, and skill development rather than routine execution. This transition requires purpose-built infrastructure: models optimized for coding and enterprise workflows, frameworks like the Agent SDK, and tools like Claude Code that help teams move from prototype to production faster. We're also finding that while coding has been the proving ground for AI agents, it's just the beginning. As agents expand into research, customer service, financial planning, and supply chain operations, the organizations that build expertise now will capture disproportionate value as the technology matures. Read the full 2026 State of AI Agents Report, here .
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
How Kepler built verifiable AI for financial services with Claude¶
来源: Claude Blog 发布日期: Apr 30, 2026 采集时间: 2026-05-28 价值评分: 9.0/10 正文字数: ~1157 字符
摘要¶
Inside a platform that indexes 26M+ SEC filings, earnings call transcripts, IR presentations, consensus estimates, and private data across 14,000+ companies and 27 global markets, and how the team behind it built AI that validates every number to the exact filing, page, and line item.
正文内容¶
Inside a platform that indexes 26M+ SEC filings, earnings call transcripts, IR presentations, consensus estimates, and private data across 14,000+ companies and 27 global markets, and how the team behind it built AI that validates every number to the exact filing, page, and line item.
In our series, How startups build with Claude , we highlight how startups are transforming their industries with AI. In this article, we share how Kepler built a trust and verification layer for AI in financial services. The quick pitch Name Kepler Founded 2025 Founders Vinoo Ganesh (CEO) and John McRaven (CTO) Stack AWS, Rust, Python, containers for orchestration Growth Indexed 26M+ SEC filings, 50M+ public documents, 1M+ private documents, and 14,000+ companies across 27 global markets in less than three months.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
Code w/ Claude London 2026: Rethinking how we build¶
来源: Claude Blog 发布日期: May 26, 2026 采集时间: 2026-05-28 价值评分: 8.2/10 正文字数: ~635 字符
摘要¶
Couldn't make it to London? Check out our Code w/ Claude London event recap and watch our session recordings.
正文内容¶
Couldn't make it to London? Keynotes and breakout sessions are now live.
This week in London, we brought Code w/ Claude to Europe. The event brought together builders, developers, and founders for two days of keynotes, breakout sessions, and workshops with the teams building Claude.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集