Leverage powerful AI coding agents to perform deep, exhaustive benchmarking on complex infrastructure problems. This workflow automates the process of testing numerous solutions, like different database engines or index types, to find the optimal configuration without manual effort.
Start with a challenging engineering problem, such as slow database queries. Define a clear, open-ended goal for the AI agent (e.g., 'make these queries faster') without prescribing the solution.
Provide a powerful coding agent like Codex with access to a realistic testing environment. This involves using production-like data and running the agent on a high-powered remote machine (e.g., AWS EC2) to handle the intense compute load. Use tools like tmux to manage multiple, persistent agent sessions.
Task the agent with exploring the entire solution space by running experiments continuously for hours or even days. For a slow query problem, this could involve instructing the agent to test every available open-source column store format, benchmark every compatible execution engine against those formats, and test different index types like bloom filters.
Leverage the advanced vision and document layout capabilities of Claude Fable 5 to generate clean, printable PDF worksheets, like handwriting practice sheets, for educational or personal use.
Use Claude Cowork to handle the entire product return process, from finding the receipt in your email to drafting a compelling refund request. This workflow saves time on tedious administrative tasks for faulty items.
Create a personal shopping filter in Claude that understands your values, vets brands for quality, and helps you avoid low-quality, drop-shipped goods. This system finds durable, well-made items by researching brand history and materials.
Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.