AI Practices & Workflows
How people use AI day-to-day. Prompt engineering, skill files, automation patterns
Master Claude Code's 1M Context Window: Rewind, Compact, Clear, and Subagents
Claude Code's 1M token context window opens longer autonomous sessions but introduces 'context rot' -- degraded performance as the window fills. Master four turn-end tools: /rewind to drop bad branches, /compact to summarize and continue, /clear to start fresh with a distilled brief, and subagents to wall off noisy work in their own context.
Claude Code Subagent Patterns: 10 Production Workflows That Actually Work
Claude Code subagents are the underused unlock for keeping a long session productive. The pattern is always the same -- delegate a noisy task to a child agent with its own clean context window, get back only the conclusion -- but the workflows people actually run with it cluster into a small number of high-value patterns. This guide walks the 10 patterns we use weekly: verifier, codebase scout, spec-to-tests translator, refactor proposer, PR reviewer, migration planner, bug hunter, docs generator, convention detective, and library briefer. Each one ships with a copy-paste prompt template and the failure mode to watch.
Claude Opus 4.7 Best Practices: How to Actually Get the Most Out of the Upgrade
Claude Opus 4.7 follows instructions more literally than 4.6, runs longer agentic tasks more reliably, and ships a new xhigh effort level. The practical effect: prompts that worked before may now produce surprises, and the workflows that earn the upgrade are the ones built around detailed plans, deliberate effort selection, auto mode with /fewer-permission-prompts, and explicit verification steps. Boris Cherny's day-of-launch tips are the honest playbook; this guide breaks them down with the workflows readers can apply this week.
Internal AI Workspaces: The Ramp Glass Playbook for Company-Wide AI Adoption
Ramp's internal 'Glass' platform pushed AI adoption to 99 percent by treating the harness -- not the model -- as the bottleneck. Every employee gets a fully configured workspace via SSO, a marketplace of 350+ shared skills, persistent memory from existing systems, and scheduled automations. Here is the playbook other companies can borrow.
Prompt Engineering Guide for Developers in 2026: Techniques That Actually Work
Effective prompt engineering in 2026 centers on three techniques: structured context (giving the AI clear project details), explicit constraints (defining boundaries and output formats), and iterative refinement (building on AI outputs step by step). These methods consistently produce better results than vague instructions.