Blog Posts

Anthropic engineering, product, and enterprise AI articles.

Article list

• 01 · Scaling Managed Agents: Decoupling the brain from the hands · 2026-04-08
• 02 · Claude Code auto mode: a safer way to skip permissions · 2026-03-25
• 03 · Harness Design for Long-Running Application Development · 2026-03-24
• 04 · Building a C Compiler with a Team of Parallel Claudes · 2026-02-05
• 05 · Quantifying infrastructure noise in agentic coding evals · 2026-02-05
• 06 · Designing AI-Resistant Technical Evaluations · 2026-01-21
• 07 · Demystifying Evals for AI Agents · 2026-01-09
• 08 · Effective harnesses for long-running agents · 2025-11-26
• 09 · Introducing Advanced Tool Use on the Claude Developer Platform · 2025-11-24
• 10 · Code Execution with MCP: Building More Efficient Agents · 2025-11-04
• 11 · Beyond permission prompts: making Claude Code more secure and autonomous
• 12 · Equipping Agents for the Real World with Agent Skills · 2025-10-16
• 13 · Effective Context Engineering for AI Agents · 2025-09-29
• 14 · A Postmortem of Three Recent Issues · 2025-09-17
• 15 · Writing Effective Tools for Agents — with Agents · 2025-09-11
• 16 · Desktop Extensions: One-click MCP server installation for Claude Desktop · 2025-06-26
• 17 · How We Built Our Multi-Agent Research System · 2025-06-13
• 18 · Best practices for Claude Code
• 19 · The "think" tool: Enabling Claude to stop and think in complex tool use situations · 2025-03-20
• 20 · Raising the bar on SWE-bench Verified with Claude 3.5 Sonnet · 2025-01-06
• 21 · Building Effective Agents · 2024-12-19
• 22 · Introducing Contextual Retrieval · 2024-09-19
• 23 · Eval Awareness in Claude Opus 4.6's BrowseComp Performance · 2026-03-06
• 25 · Introducing Claude Opus 4.8 · 2026-05-28
• 26 · How we contain Claude across products · 2026-05-25
• 27 · KPMG integrates Claude across its core business and workforce of more than 276,000 in strategic alliance · 2026-05-19
• 28 · Anthropic acquires Stainless · 2026-05-18
• 29 · PwC is deploying Claude to build technology, execute deals, and reinvent enterprise functions for clients · 2026-05-14
• 30 · Bringing MCP 2026-07-28 to Claude · 2026-07-28
• 31 · The 2026-07-28 Specification · 2026-07-28
• 32 · How the product designer who built Claude Design uses it to explore ideas before building them · 2026-07-24
• 33 · Claude models explained: choosing the best model for your use case · 2026-07-24
• 34 · The new rules of context engineering for Claude 5 generation models · 2026-07-24
• 35 · Think through hard problems in voice mode · 2026-07-23
• 36 · Four role-based certifications for the people who put Claude to work for customers · 2026-07-23
• 37 · Building verification loops in Claude Code with skills · 2026-07-22
• 38 · How Outtake built a cyber investigator on Claude · 2026-07-22
• 39 · How Datadog built a “universal machine tool” for Claude Code · 2026-07-21
• 40 · How Anthropic secures its AI-native software development lifecycle · 2026-07-21
• 41 · Working at the frontier: How Rakuten builds agents overnight with Claude Fable 5 · 2026-07-20
• 42 · Zero risk isn't the job: a CISO's guide to agentic AI · 2026-07-17
• 43 · Working at the frontier: How Cursor knew Claude Fable 5 was ready for the hardest 1% of problems · 2026-07-17
• 44 · Working with Claude Fable 5 in Claude Cowork · 2026-07-16
• 45 · How Anthropic runs large-scale code migrations with Claude Code · 2026-07-16