Boris Cherny

Anthropic launched Claude Cowork, a general-purpose desktop AI agent for non-technical users that extends Claude Code capabilities to productivity tasks, competing directly with Microsoft Copilot in enterprise markets.

“Head of Claude Code who discussed how Anthropic built Cowork in approximately a week and a half using Claude Code itself.”

“Lead of Claude Code at Anthropic who described how engineering, product, design and data science roles are merging into one with AI agent support.”

“Internal Anthropic employee using Claude Duet to manage his mental to-do list during walks.”

“Created Claude Code and emphasized importance of giving Claude a way to verify its work”

“Engineer leading Claude Code at Anthropic describing how engineers spend time prompting models, talking to customers, and deciding what to build when Claude writes their code.”

“Claude Code creator revealing he runs 5-10 Claude agents in parallel on multiple tabs for productivity.”

“Anthropic researcher cited for claiming frontier labs have largely solved prompt injection threat despite benchmark evidence otherwise.”

“Team lead at Anthropic who built Claude Code to seize the opportunity when model capabilities exceeded harness capabilities.”

“Anthropic engineer addressing complaints about Claude Opus 5 verbosity and explaining optimization strengths of the model.”

“Creator of Claude Code who recommends deleting config and skills files every 6 months to optimize system performance with newer models.”

“Head of Claude Code reporting on Anthropic's advances in training models to resist indirect prompt injection attacks.”

“Head of Claude Code at Anthropic stating the team has used auto mode exclusively and couldn't imagine going back to permission prompts.”

“Discussed how Opus 5 achieves unprecedented capabilities, explained the deletion of system prompts, and detailed how Claude Code is rebuilt with each model generation.”

“Claude Code creator who deleted over 80% of Claude Code's system prompt for Opus 5, discussing model capabilities and ablation testing.”

“Signed a statement with other AI industry leaders calling for government regulation of automated AI development.”

“Stated he doesn't prompt Claude anymore but uses loops running that prompt Claude.”

“Anthropic engineer urging teams to embed institutional rules directly into codebases for AI agents using lint rules and CI pipelines.”

“Shared tips on how to get the most out of Opus 4.7 model.”

“Built Claude Code and stated he no longer prompts Claude manually, instead engineering autonomous loops.”

“Creator of Claude Code who pioneered loop engineering paradigm for autonomous AI coding.”

“Engaged with Rahul Sengottuvelu's post on AI code review practices and risk management.”

“Explained Anthropic's reasoning for restricting Claude subscriptions from third-party agent frameworks due to unsustainable compute costs.”

“Head of Claude Code at Anthropic who announced restrictions on third-party tools like OpenClaw due to usage patterns and profitability pressures.”

“Creator of Claude Code at Anthropic who agreed with Andrej Karpathy about feeling behind with new AI tools and noted that new graduates may fare best in this environment.”

“From same Odesa town as Lenny Rachitsky; discovered to share hometown during Rachitsky podcast.”

“Boris Cherny led live coding sessions and spoke about Claude Code's capabilities at Code with Claude SF event.”

“Explained the technical reason for OpenClaw cutoff: cache hit rates and token processing costs differ between first-party and third-party tools.”

“Conducted a rare public livestream showcasing his authentic everyday workflow using Claude Code, marking the first time an AI coding tool creator has offered an unfiltered, real-time demonstration.”

“Creator and Head of Claude Code at Anthropic; discovered to be from the same town in Ukraine as Lenny Rachitsky.”

“Boris Cherny is part of the team that built Claude Code.”

“Claude Code lead who responded to nerf allegations by discussing adaptive thinking, effort defaults, and thinking visibility.”

“Said prompt cache misses with 1M token context window are expensive and Anthropic is investigating 400,000-token default context window.”

“Announced that Claude subscriptions will no longer cover usage on third-party tools like OpenClaw due to engineering constraints.”

“Announced that Claude subscriptions will no longer cover usage on third-party tools like OpenClaw due to engineering constraints.”

“Boris Cherny announced Anthropic's update to Claude subscriptions regarding third-party tool access and usage billing changes.”

“Explained Anthropic's decision to change pricing for third-party harnesses, stating subscriptions weren't built for their usage patterns.”

“Said the DMCA takedown was accidental and retracted bulk of the notices, limiting takedown to intentionally leaked repositories.”

“Stated 'We're going to start to see the title of software engineer go away. It's just going to be builder or product manager.'”

“Anthropic engineer who created Claude Code stated that vibe coding works best for prototypes or throwaway code, not mission-critical software.”

“Boris Cherny, the father of Claude Code, has stated that 100% of his code is written by Claude Code itself.”

Anthropic's founding engineers of Claude Code share practical usage tips including plan mode for complex tasks and shared settings files for team standardization, revealing how the tool has transformed team productivity.

“Co-creator of Claude Code who discussed practical tips and philosophy for developing the agent on the AI & I podcast.”