n8n vs Custom Code for Engineering Automation: The Decision, and the Bug That Proved It Right

Description and use

Choosing between n8n and custom code for automating content publishing. Silent failures in the Buffer API were caught in custom code, where debugging is straightforward, whereas n8n could hide such issues.

Problem

Choosing between a no-code workflow tool (n8n) and custom code for automating content publishing, especially when silent failures can occur without alerts.

Outcome

Custom code allowed transparent debugging and code reviews, preventing silent failures that a no-code canvas would obscure.

Workflow

• 1. Capture post approval via Telegram bot.
• 2. Queue post for publishing in custom Python pipeline.
• 3. Call Buffer API to publish.
• 4. Mark post as published in queue.
• 5. Monitor for errors and alert if any occur.

Evidence

The Buffer API adapter treated an error response as a normal one, never checked the status, so the X post never actually went out while the pipeline marked it published.

Source

I genuinely don't have a mental model for debugging that failure mode there, a canvas doesn't hand you the same thing a stack trace and git blame hand you in code.

Source