Jayson Bautista.

AI email management.

An inbox workflow with classification, draft review, reminders and reporting.

  • n8n
  • Gmail
  • Groq
  • Google Sheets
  • Telegram

Personal project · Source-backed architecture

The problem explored

Keeping classification, reply review and follow-up connected across an inbox.

An inbox workflow with classification, draft review, reminders and reporting. The supplied diagrams document the configured connections, rather than verified production outcomes.

Authentic classification and draft-routing excerpt from my AI email management n8n project
AI email managementSelected branch from the source diagram
Full source

Scroll sideways to inspect the diagram. Zoom enlarges labels; Reset zoom restores this preview.

Architecture summary · read alongside the source

  1. Receive & classify
  2. Route & draft
  3. Review & send
  4. Follow up & report

Inside the processing logic

Receive & classify

  • A Gmail Trigger feeds an AI Agent Classifier connected to Groq.
  • Parsed output reaches label lookup and label-existence checks.
  • The diagram includes creating a label and adding a label to a message.

Route & draft

  • A sender lookup checks whether the sender is a VIP.
  • VIP and high-priority branches contain AI draft generation.
  • Draft details are logged in Google Sheets and notifications go to Telegram.

Review & send

  • Telegram Callback Query starts a separate reply-review path.
  • The path retrieves draft details before the Gmail send-message node.
  • Status updates and a Telegram confirmation follow the send step.

Follow up & report

  • A reminder branch reads unanswered emails and checks for overdue items.
  • Other branches show daily summaries, weekly reports and dashboard metrics.
  • A spam-checking branch and an unsubscribe-analysis branch are also visible.

Step through the explanation

Dennice, an original illustrated robot guide, walks through the diagram. Start with Next, or choose autoplay. Captions and a full transcript are included; sound is optional.

Open the film on its own

Source evidence

These are the supplied local files used to write this explanation. Node labels and connections are visible; exported workflow JSON and execution logs were not supplied.

AI email management source diagram 1
Source: automation1/1.png. Open original image

AI email management source diagram 2
Source: automation1/2.png. Open original image

AI email management source diagram 3
Source: automation1/3.png. Open original image

Visible outputs

Gmail labels and drafts, review-driven replies, Telegram updates and Sheets logs.

These outputs are represented by nodes in the source. Delivery success and business results require execution evidence.

What this case study does—and doesn't—verify

The screenshots document node names, integrations and connections. Prompts, credentials, exact routing rules and execution logs are not supplied. Sending accuracy, response times and time savings have not been measured here.

Next validation steps

Before operational use, I would inspect the workflow export, confirm credentials and permissions, test representative inputs and error paths, and review the outgoing messages with the intended user. These are proposed checks, not completed workflow execution tests.

Next: Resume screening

Let’s make
something work.

Have a development role, an automation idea or a website project in mind? I’d like to hear about it.

Jayson Bautista · Automation Developer & Web Developer