Boris Cherny's AI Coders: A Glimpse Into The Future of Software Engineering
Boris Cherny, architect of Claude Code, is shaking up software development with an innovative workflow that treats AI as a team, not just a tool. Here's why that matters.
Boris Cherny, the brain behind Claude Code at Anthropic, has captured the tech world's attention. His recent disclosure of a unique AI-driven coding workflow is setting the stage for a new era in software development.
AI's New Role: Not Just an Assistant
Forget traditional coding loops. Cherny operates with five AI agents simultaneously, resembling a fleet commander's approach rather than a coder's. Five Claudes run in parallel within his terminal, each handling different tasks, from running test suites to refactoring legacy modules. It's not just multitasking. it's a symphony of coordinated AI efforts.
This method is more than a gimmick. It's reshaping how developers think about productivity. Why limit AI to being an assistant when it can be a full-fledged team member? That's the real question programmers need to ask themselves.
Opus 4.5: Slow and Smart Wins the Race
In a move against the industry's obsession with speed, Cherny opts for Anthropic's Opus 4.5 model. Yes, it's slower, but it's smarter and requires less human intervention. In the end, it saves more time by reducing errors, a essential insight for tech leaders. The numbers tell a different story: faster isn't always better AI.
This approach aligns with Anthropic's 'do more with less' philosophy. While competitors invest heavily in infrastructure, Cherny proves that intelligent orchestration can achieve exponential gains.
Learning from Mistakes: A Smarter AI
AI amnesia is a common hurdle, but Cherny tackles it head-on with CLAUDE.md. This shared file ensures that every mistake becomes a learning opportunity, transforming the codebase into a self-correcting entity. The longer his team works with Claude, the more efficient it becomes. It's a fascinating blend of machine learning and human insight.
Automation: Cutting Through the Noise
Automation is at the heart of Cherny's workflow. Using slash commands and subagents, he drastically reduces repetitive tasks. A simple command like/commit-push-prstreamlines the entire version control process. Meanwhile, specialized subagents handle specific phases, ensuring a smooth development lifecycle.
The verification loop is another big deal. Claude doesn't just generate code. it tests and verifies it, enhancing quality dramatically. This isn't just automation. it's a revolution in software engineering.
The takeaway? The future of coding is here, and it's powered by AI. Those who embrace this change won't just be more productive. they'll redefine what it means to be a programmer. Will you be typing, or commanding?
Key Terms Explained
A mechanism that lets neural networks focus on the most relevant parts of their input when producing output.
A branch of AI where systems learn patterns from data instead of following explicitly programmed rules.
An AI safety company founded in 2021 by former OpenAI researchers, including Dario and Daniela Amodei.
Anthropic's family of AI assistants, including Claude Haiku, Sonnet, and Opus.