Y Combinator• Feb 17, 2026• 1:33:28Interview

Inside Claude Code With Its Creator Boris Cherny

From Y Combinator

Boris Cherny

Executive Summary

Boris Czerny, creator of Anthropic's Cloud Code, discusses the product's accidental origins as a simple CLI tool and its evolution driven by latent user demand.
He outlines a core development philosophy: build for the AI model of six months from now, not for today's capabilities.
Czerny reveals that Cloud Code has driven a 150% productivity increase per engineer at Anthropic, with 70-90% of code now written by the AI.
He predicts the eventual obsolescence of the 'software engineer' title and the rise of multi-agent 'swarms' capable of building entire features autonomously, a process already used to create Cloud Code's plugin system.

8 quotes

Concerns Raised

Potential for catastrophic misuse of advanced AI models (e.g., designing bioweapons, zero-days)
Reaching AI Safety Level 4 (ASL4), where models become recursively self-improving, requires extreme caution before release

Opportunities Identified

Massive (1000x) developer productivity gains through AI coding agents
The ability for AI agents to build entire software features autonomously
The democratization of software development, enabling non-coders to build complex applications
Rapid product iteration by using AI to generate dozens of prototypes in hours instead of weeks

Key Themes

Building for Future Models

The core strategy behind Cloud Code is to anticipate and build for the capabilities AI models will have in six months, rather than being constrained by current limitations. This means accepting that today's product might not be fully functional but will become powerful as the underlying models improve.

This provides a strategic framework for founders in the AI space, emphasizing long-term vision over short-term performance hacks that will become obsolete.

Latent Demand Driven Development

Cloud Code's features, including its CLI-first interface and 'plan mode', were not part of a master plan but emerged from observing how users were already trying to solve problems. The product's success comes from making existing workflows easier, not inventing new ones.

This reinforces the classic product principle of 'talk to your users' and highlights its heightened importance in the fast-moving AI landscape.

The Ephemeral Nature of Code

The codebase for Cloud Code is in a constant state of flux, with no code from six months ago still existing. Scaffolding built to compensate for model weaknesses is treated as temporary tech debt, to be deleted when the next model generation makes it redundant.

This signals a major shift in software engineering, where the value lies in directing the AI, not in the longevity of the code itself.

The Rise of Autonomous Agent Swarms

Complex software development is moving towards multi-agent systems. Cloud Code's plugin feature was built entirely by a 'swarm' of agents coordinated via an Asana board, demonstrating a new paradigm for autonomous software creation.

This points to the future of engineering management and project execution, where tasks are delegated to teams of AI agents rather than human engineers.

The 1000x Engineer and Role Convergence

AI tools are creating unprecedented productivity leaps, with Anthropic seeing a 150% increase per engineer. This is blurring roles, with designers, PMs, and finance staff now coding, and leading to the prediction that the 'software engineer' title will fade in favor of more generalist 'builders'.

Professionals should anticipate a convergence of roles where technical implementation skills are table stakes for everyone, not just a specialized function.

Get started free

Topics

Cloud Code Anthropic AI Agents Developer Tools Software Development Productivity LLM Applications Future of Work AI Safety Product Management CLI TypeScript Prompt Engineering Multi-agent systems

Processed Feb 24, 2026 yt-dlp + mlx-whisper + Gemini