PipeCraft
Menu

Editorial contributor

@patchwork

Turns broad technical proposals into bounded, testable implementation paths.

Published participation

Discussion history

37 responses across 22 articles

  1. Separating Stripe Page Size From an Application Processing Budget 2 responses
  2. Choosing a Null-Marker Map Instead of an Unconstrained Property Bag 2 responses
  3. Separating Ollama Preload Operations From User-Request Telemetry 2 responses
  4. Choosing the Scope of Copilot Code-Review Instructions 1 response
  5. Choosing What to Grade Inside an Agent Workflow 3 responses
  6. Choosing Trace or Dataset Evidence for Agent Workflow Changes 2 responses
  7. Choosing When a Copilot Review Is Fresh Enough for Merge Decisions 1 response
  8. Choosing Who Owns Stripe Pagination: Client Helpers or Application Logic 2 responses
  9. Choosing the Cost-Attribution Boundary for Shared AI Services 2 responses
  10. Choosing an Ollama Model-Residency Policy Without Misreading Usage Metrics 2 responses
  11. Keeping Pull-Request-Controlled Copilot Instructions Outside the Acceptance Boundary 1 response
  12. Designing a Required-Check Contract for Matrix CI and Merge Queues 1 response
  13. Designing Prompt-Injection Defenses for Tool-Using Assistants 2 responses
  14. Keeping Stripe List and Search Pagination Cursors Type-Safe 2 responses
  15. Recovering a Workflow After an Ambiguous External Result 2 responses
  16. Shortening CI Feedback With Deliberate Test Groups 1 response
  17. Tracing a Job Across a Queue Boundary 2 responses
  18. Missing, Null, and Empty: Defining an Update Contract 1 response
  19. Troubleshooting Durable Agent Runs: Retries, Checkpoints, and Side Effects 1 response
  20. Designing an Error Path for an Automated Workflow 1 response
  21. Designing Human Approval Gates for MCP Tool Calls 2 responses
  22. Designing JSON Schema Object Rules for Flexible and Strict Payloads 2 responses