Claude

Evals for taste: Hill-climbing a slide-generation agent

Agents that remember

Building persistent memory for AI agents requires moving beyond isolated, ephemeral sessions toward integrated architectures. Cloud Managed Agents address this by implementing memory stores—persistent, file-system-like resources that enable agents to read and write information across multiple interactions. To prevent t...

How we Claude Code

Making agentic workflows trustworthy and verifiable with a custom DSL

Fighting financial crime with Claude Cowork

Where code meets court: AI at the legal-technical frontier

How AirOps chases friction to build AI products with Claude

How Metaview built self-improving prompts for application review

Teaching agents to learn from your team

Beyond the basics with Claude Code

Running an AI-native engineering org

Getting more out of the Claude Platform

The prompting playbook

Prompt engineering requires a systematic approach centered on rigorous evaluations to track performance and identify regressions during model migrations. Maintaining production prompts involves applying general hygiene, such as using XML tags for structure and removing redundant, legacy instructions that cause overfitt...

The capability curve

Designing with Claude: From prompt to production

Anthropic Labs developed the "Claude Design" product in just 10 weeks by operating as a "bet factory" that prioritizes rapid iteration over traditional planning. Instead of relying on formal Product Requirement Documents (PRDs), the team utilized prototypes and daily user feedback to guide development. This lean approa...

Building the best agentic analytics harness: Powered by Claude, built with Claude Code

Omni leverages Claude AI to power its analytics platform, transforming natural language queries into precise SQL through a robust semantic layer. The engineering team utilizes Claude Code to accelerate development, maintaining high velocity by embedding context directly into data definitions. The company’s AI agent, "B...

Building with Claude on Google Cloud

Building signals that trade themselves

Picking the right model

Memory and dreaming for self learning agents

Anthropic’s new "Memory" and "Dreaming" features for Claude Managed Agents enable autonomous systems to learn and improve across tasks, sessions, and multi-agent environments. Memory functions as a persistent, file-system-based storage layer that allows agents to carry forward strategies and avoid past mistakes. Comple...