Claude Code Patterns for Big Codebases That Scale
Reusable patterns for Claude Code in large repos: spec-shaped prompts, plan gates, tool scoping, packaged context, and self-verifying loops that scale.
7 min read
Agentic AI & LLMs
Agentic AI & LLMs archive page 67 of 631
Agentic AI & LLMs
Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
Latest articles
Reusable Claude agent patterns: prompts, tools, context
Code-level patterns for prompts, tools, and context in Claude agents — reusable building blocks AI-native founders copy across every product.
7 min read
Build your first Claude agent: a step-by-step guide
A concrete, engineer-followable walkthrough to build a production Claude agent: loop, tools, MCP, memory, and evals — for AI-native founders.
8 min read
Claude Code in a Monorepo: A Step-by-Step Walkthrough
A reproducible walkthrough for using Claude Code on a large monorepo: memory files, task scoping, blast-radius mapping, sliced edits, and safe shipping.
7 min read
AI-native startup architecture with Claude, end to end
The end-to-end architecture of an AI-native startup on Claude: agent loop, context assembly, tools, MCP, memory, and evals — a founder's stack map.
8 min read
How Claude Code Works Inside Large Codebases
Inside Claude Code's architecture for big repos: the agent loop, context strategy, repo navigation, subagents, and how the pieces fit end to end.
7 min read
Where Claude Computer Use Is Heading Next
Where Claude computer and browser use is going next — MCP tools, multi-agent orchestration — and concrete steps to prepare your team and code today.
6 min read
How to Measure Claude Computer-Use Success
The metrics and signals that prove a Claude computer-use or browser agent works — task success, intervention rate, cost per outcome, and eval gates.
6 min read
A Real Claude Browser-Use Build, Start to Finish
One Claude browser-use automation from a messy business problem to a shipped, supervised outcome — decomposition, dead ends, and what made it stick.
6 min read