Writing about web development, AI, and software engineering.

Jul 10, 2026·7 min read

Why the Most Expensive Models Will Cost You Less

Benchmark data from Anthropic and Artificial Analysis shows that frontier models at reduced effort settings repeatedly beat cheaper models at maximum effort on both quality and cost per task. Per-token pricing is the wrong selection criterion.

Feb 6, 2026·9 min read

Agent Teams in Claude Code: How They Actually Work

How Claude Code's agent teams turn parallel work into collaborative work. Communicate with teammates directly, steer the orchestrator, and coordinate complex tasks without losing your mind or your context.

Dec 3, 2025·4 min read

The Case for Task-Aware Reasoning Budget Allocation

Research from DeepSeek, Berkeley, and Google DeepMind shows that reasoning budgets must be allocated based on task structure, not maximized by default. Here's the evidence and a routing architecture.

Oct 17, 2025·7 min read

Using Sub Agents to Save Context Window or Save Time

Discover context management in Claude Code with sub agents. Understand when to utilize parallel or sequential execution, avoid context bloat, and coordinate multi-agent workflows for complex development workflows.