Freelancer - n8n Automation Expert – Gmail / Google Sheets / Slack / AI Email Classification

Jobs
1

Resolved translation direction confusion and proceeded with task

We are looking for an experienced automation developer to build 6 interconnected email automation workflows for a B2B travel agency. A fully detailed requirements document is already in place – this is a build job, not a discovery project.

The Client: A German-speaking travel agency. Primary project language is German. English communication is fine on our side.

The Project – 6 Workflows:

  • Automatic Email Labeling – AI-powered classification of incoming emails into 5 categories (invoices, hotel updates, accounts (approx. 50 different clients), urgent, booking changes) via Gmail API + LLM of your choice (Claude, OpenAI, Gemini)

  • Hotel Update Forwarding – Auto-forward labeled hotel updates to the sales team inbox

  • Super Urgent Escalation – Keyword detection + automatic Slack notification to sales channel

  • Request Status Tracking – Automatic intake of new requests into Google Sheets with unique reference number, full timestamp tracking across all status stages (Pending → Acknowledged → Offer Sent → Follow Up → Negotiation → Confirmed/Declined)

  • Daily Reporting – Formula-based dashboard in Google Sheets + daily Slack digest with link and one-line summary

  • Incoming Confirmation – Auto-reply to first-contact emails only, with auto-reply loop protection

Tech Stack at Client:

  • Gmail (Google Workspace) – primary email platform

  • Google Sheets – tracking & reporting database

  • Slack – internal communication & alerting

  • Proprietary booking engine (REST API access to be clarified)

  • LLM API for multilingual email classification (DE/EN/FR)

Technology:

We are open to different approaches. Our preference is n8n (self-hosted, installed by us) as the primary automation layer – but we are open to alternatives if you have strong experience with another solution that makes more sense for this project.

For the tracking database, Google Sheets is our first idea, but we are also open to using a proper database if you can make a strong case based on your expertise.

This will be the first set of automations for the client – more projects are already in the pipeline. :blush:

Required Skills:

  • n8n or comparable automation platform (advanced level)

  • Gmail API / Google Workspace API

  • Google Sheets API

  • Slack API

  • LLM/AI API integration (Claude, OpenAI or Gemini)

  • Clean workflow documentation & error handling

  • REST API integration experience

What we provide:

  • Fully detailed requirements document with all fields, status stages and logic pre-defined

  • Direct contact person throughout the project

What we expect:

  • Structured proposal with fixed rates per workflow and as a total package

  • At least 1 example of a similar project (email automation, Gmail integration or Sheets tracking)

  • Clean, documented workflows – no black-box builds

  • German or English communication

6 Likes
2

Hey :waving_hand:,

I’m Milan, with 8 years of experience in Business Automation and AI. Including 2 years at Apify working on enterprise-level browser automation.

Currently specializing in n8n, but also proficient in Python & Javascript.

Find out more about my work here:

If you think I might be a match, please:

Book a call here with me

Or reach out at hello@smoothwork.ai

Looking forward to hearing from you!

2 Likes
3

Hey there! :waving_hand:

This is a really well-scoped project. I’d have upfront: are you leaning toward self-hosted n8n for data sovereignty reasons, or is that flexible? Just want to understand if there are compliance/infrastructure constraints I should factor into the proposal.

On the tracking layer: Google Sheets can absolutely work here, especially with the unique reference numbers and status stages you’ve outlined. That said – if you’re doing daily syncs and the workflow volumes are high, a lightweight database (Postgres + n8n connectors) might reduce API throttling headaches down the line. Happy to make that case more formally in the proposal if it’s worth exploring.

What I’d need to move forward:
• A quick look at the requirements doc (or at least the email categorization logic and the status stage workflows) so I can scope accurately
• Clarification on the proprietary booking engine API (is it documented?)
• Timeline expectations (is this build-in-parallel or one after another).

I’ve built similar projects: Gmail classification with Claude, status-tracking in Sheets, Slack escalation workflows.

Happy to share examples once we’re in talks.

1 Like
4

If you need I am available here:

I am one of top 50 verified template creators:

I am also capable of building custom community n8n node too:

  1. https://www.npmjs.com/package/n8n-nodes-pdfbro
  2. https://www.npmjs.com/package/n8n-nodes-ocrbro
  3. https://www.npmjs.com/package/n8n-nodes-ttsbro

Apart from that I’m also a full-stack developer with the right Gen AI experience, which makes me a solid plus for your team[but right now only vibecoding]

Check my recent gen ai projects… I built a native Android automation agent too. It’s worth a look:

I can build complex AI automations directly in code, not just inside n8n

I recently started posting my n8n work on YouTube with explanations:

You can schedule a quick call with me: N8N Project Consultation with Vaar | Iamvaar | Cal.com

Fun fact… I even made an n8n workflow to find a few n8n project leads for myself so I truly believe in what I do…

9bSkKu
9bSkKu1220×1470 122 KB

I started asking My recent clients for honest feedbacks so here is one testimonial: https://www.youtube.com/watch?v=TqBy3SVCHgQ&list=PLAJltY5bp6yiZ3sFBjm7bfrkLXSGtJX8m

Here is my linktree: iamvaar | X | Linktree

And I am also capable of building gohighlevel automations and other CRM’s too, here are my recent gohighlevel integrated automation videos:

  1. https://youtu.be/eZ3t7PfXFho
  2. https://youtu.be/nSJzOB9JZlY

And I also built low latency voice appointment scheduler with live ai avatar[in code]:

I built AI search visibility tracker[I am capable of building complex web scraping automations in python too]:

And done many other big projects too which are under NDA’s so technically I can’t reveal them.

5

Hi! This project is right in my wheelhouse. I’m an automation developer with 3+ years of experience building exactly this kind of stack: n8n, Gmail API, Google Sheets, Slack, and LLM integrations (Claude/OpenAI) for multilingual workflows.

I’ve handled AI-powered email classification, Slack alerting systems, and Sheets-based tracking pipelines in past projects, so all 6 workflows here are very familiar territory for me.

I’m also a Top Community Member on Fiverr with a strong track record of clean, documented builds no black-box work.

https://www.fiverr.com/sellers/moosakhuwaja/

Happy to jump in as soon as you share the requirements doc. Feel free to reach out!
Here is my Personal Whatsapp Number: +923219283372

Best regards,

6

Hey — great question.

I’m comfortable working with self-hosted n8n, and I’m assuming that preference is mainly for data control and reliability — which makes sense for a setup like this. I can fully adapt to that.

On the tracking layer: Google Sheets will work for the initial phase, especially with the structured status flow you’ve defined. That said, if you’re expecting higher email volume or more complex state transitions, I’d recommend moving to Postgres early to avoid API limits and keep the system stable long-term. Happy to scope both options depending on your expected load.

To move forward, I’d need:

  • Access to the requirements doc (especially classification logic + status workflows)

  • Details on the booking engine API (documentation, endpoints, auth)

  • Expected email volume per day

  • Preferred timeline (parallel build vs phased rollout)

I’ve built similar systems with Gmail + AI classification, Sheets-based tracking, and Slack escalation workflows — can share relevant examples once we align on scope.

Portfolio (with demos):
https://muhammad-ai-automations.notion.site/Muhammad-Bin-Zohaib-AI-Automation-Projects-29da292a241380f889c2e337a134c010

If this aligns, I can put together a fixed-scope breakdown per workflow and total package.

7

Hi!

I’ve looked through the 6 workflows. This is a classic case where n8n + Gemini/Claude can save the team hours of manual work. I’m an automation engineer, and I specialize in exactly this kind of “smart” routing.

Here’s how I’ll build this for you:

  • Reliable AI Labeling: I don’t just send a prompt; I use JSON mode to make sure the AI always gives back a clean category that the Gmail API can actually use. No errors, no “hallucinations.”

  • Bulletproof Logic: For the auto-replies, I’ll set up a proper “loop protection” so you don’t end up with two bots talking to each other forever.

  • Clean Data: I can definitely start with Google Sheets, but I’ll build it so that if you decide to move to a “real” database later, we can switch in an hour without rebuilding everything.

  • Documentation: I hate “black boxes” as much as you do. Every workflow will be commented and easy to manage after I’m done.

Why me? I’m a developer (JavaScript/Python), so if a standard node doesn’t work with your booking engine, I’ll just write a custom script to fix it.

My Portfolio: https://mikedevai.netlify.app/

Rates & Availability: I’m ready to give you a fixed price for the whole pack once I see the requirements doc. I work fast and I’m available for a call this week.

Contacts:

Let’s build something solid!

8

Hi BenKo,

Saw your post in the n8n community about building the 6 interconnected email automation workflows for the B2B travel agency.

This is a very well-structured automation system — especially the combination of AI email classification, status tracking in Google Sheets, and Slack-based escalation/reporting workflows.

I’d be happy to collaborate on this.

I specialize in building production-grade automation systems using n8n with Gmail, Google Sheets, Slack, and LLM-based classification layers. Most of my work focuses on structured email processing pipelines, multi-step workflow orchestration, and API-driven business automation.

From your breakdown, this is essentially an AI-powered email operations system across Gmail + Sheets + Slack, with LLM acting as the routing and decision layer across 5+ business-critical workflows. That aligns closely with systems I’ve built involving inbox classification, status tracking pipelines, and automated reporting systems.

I’ve worked on similar implementations involving:

• Gmail API + AI email classification (multi-category routing systems)

• Google Sheets-based workflow tracking with structured status stages

• Slack escalation + alerting systems for urgent workflows

• Multi-step automation pipelines for CRM-style operations

Relevant examples:

• AI Voice Agent Automation (API + workflow integration)

https://www.upwork.com/freelancers/~0122761e4734295f4b?p=2038586338272239616⁠�

• Multi-Channel Automation Workflow

https://www.upwork.com/freelancers/~0122761e4734295f4b?p=2039118619839795200⁠�

• Shopify Order Process Automation System

https://www.upwork.com/freelancers/~0122761e4734295f4b?p=2039336822638325760⁠�

I’d be open to a quick 10–15 minute call to understand how you want to structure the 6 workflows and contribute to Phase 1 implementation.

Looking forward to hearing from you.

Best,

Folafoluwa Stephen

Email: folafoluwaolaneye@gmail.com

Click here to book a al walkthrough call session.

9

Hey Benko,

This looks like a really solid and well-structured project, exactly the kind of automation work I enjoy building.

I’m currently working with a few clients on similar systems around Gmail, Google Sheets, Slack and LLM APIs (OpenAI / Claude), especially where email classification, tracking flows and internal alerts need to run reliably together.

I mainly work as an automation engineer, building these workflows end-to-end, and n8n is usually my go-to for setups like this because it keeps everything flexible and easy to maintain.

I’m based in the EU and I speak EN / HU / RO fluently, so communication is easy. Happy to go through your requirements doc and break things down workflow by workflow if that helps.

You can check a bit more here → www.callera.ai
Or book a quick call here → https://cal.com/callera-discovery/discovery-call

Insta www.instagram.com/aibusinesshack

Thanks,
David

10

Hi BenKo,

B2B travel, six email workflows, multilingual classification across DE/EN/FR, loop-protected auto-replies. The “no black-box builds” line is the discipline I design around.

What I’d bring specifically:

  • Schema-validated AI output at every LLM boundary (Pydantic or JSON Schema) so a misfired classification is rejected, not silently written to Slack
  • An append-only audit log that records prompt, model, and decision for every email processed — so when Ops asks “why did this one get escalated?” you can show them
  • Self-hosted n8n on your own infra, with config-driven rate limits so the hotel-update forwarder cannot accidentally flood your inbox

Reference repository I maintain:

A tiered multi-agent system with the exact audit / schema-validation pattern I would apply here. CI-green on Linux and Windows, 21 tests, MIT licensed.

If that scope maps to your plan, reply and I’ll send the proposal with per-workflow fixed rates and a total-package price.

Priyanshu Kumar
AI & Automation Engineer

https://www.linkedin.com/in/priyanshu-axiom

11

Hi guys, thanks for all your great applications and sharing your experience. We are currently reviewing the applications. And will reach out if we feel its a match. If it not works out this time, we have you in mind for the next project.

We dont take further applications into account at the moment. Thanks.

Best regards, Benjmain

2 Likes
12

Hi! This job post caught my eye because I literally just finished building an almost identical architecture for a B2B consulting agency.

I work entirely on self-hosted n8n and my main focus is on system stability and data integrity. Since you already have the logic pre-defined, here is how I would handle your core technical bottlenecks:

  • Email Classification & Routing: Instead of messy multi-node setups, I use a single-node LLM strategy (OpenAI/Gemini) forcing structured JSON outputs. This makes routing to Slack or forwarding hotel updates completely bulletproof.

  • Request Status Tracking: The biggest issue with Google Sheets as a database is row duplication. I solve this by using the native Gmail Message-ID as an absolute primary key. Whether it’s Pending, Offer Sent, or Confirmed, n8n will always update the exact correct row.

  • Auto-Reply Loop Protection: I build strict trigger-level filters (e.g., excluding self-sent emails) and “Bouncer” IF nodes right at the webhook level to ensure auto-replies never trigger an endless loop.

  • Error Handling: No black-box builds. I set up idempotent nodes with auto-retry, backed by a separate Global Error Handler workflow. If an API request crashes, it instantly pings a Slack digest with the exact workflow and node name.

I’m very comfortable working with your requirements document and can break down the fixed rates per workflow.

Would love to drop you a DM or jump on a quick screen share to show you how the deduplication and error-handling logic works in my current production build! Let me know.

13

Hi BenKo,
Seasoned dev of 3 years with n8n course level 2.

I’ve built similar Gmail → Sheets → Slack pipelines for B2B clients before, so this is familiar ground — the classification layer I’d handle with Claude API using structured JSON outputs to keep labeling deterministic across DE/EN/FR, with loop protection on the auto-reply handled via label detection cross-referenced against reference numbers to avoid firing on re-opens. One thing worth flagging: with 50 concurrent client accounts, Google Sheets can miss writes under load — happy to build it on Supabase instead with Sheets as a read layer, or stick with Sheets if you prefer to keep it simple. All 6 workflows, fully documented, error handling included I can give a quote after hearing the details Ready to start on receipt of the requirements doc.

— Aryan

pmediaaryan@gmail.com

14

Hey — read through this carefully. A few thoughts before I respond formally.

The 6 workflows here are straightforward in isolation. But the way they’re interconnected means the build order matters more than most developers will mention. WF1 (AI Labeling) isn’t just the most complex — it’s the load-bearing wall. If it misfires, WF2, WF3, and WF6 all break silently. Most developers will build these in sequence and hand them over. The problem shows up 3 weeks later when a hotel update gets routed to the wrong inbox and nobody notices.

The part I’d want to nail down before quoting: the 50 client account mapping in WF1. There are three ways to handle that — each with a different failure mode depending on how often client names change. The requirements doc will tell me which approach fits.

WF6 also has a detail that’s easy to underestimate. “First-contact only” auto-replies are simple. Auto-replies with loop protection that still works when the same client sends a second email next week — that’s the part that breaks in production if not designed correctly from day one.

I work primarily in n8n, with Gmail API, Google Sheets, Slack, and Claude/OpenAI for the LLM layer. Built similar email classification + Sheets tracking systems before. Documentation is part of every delivery — not an afterthought.

I’m not going to send a generic quote without seeing the requirements document first. Could you share it? I’d rather give you an accurate fixed-price breakdown per workflow than a number I have to revise later.

Best
Hamza
itsameerhamza203@gmail.com

15

Hello there!

I work with n8n as my main automation tool and have hands-on experience with API integrations, LLM workflows, Telegram bots, crm automation, validation, deduplication, logging, and error handling. From your description, the Gmail → AI classification →Slack escalation → Google Sheets tracking flow is exactly the kind of operational system I can build.

Relevant background:

  • built LLM ticket triage workflows with classification, routing, draft generation, and structured logging

  • built modular n8n systems with sub-workflows, retries, fallback branches, and idempotent logic

  • worked with Google Sheets as an ops layer and with messaging systems as an alerting interface

  • deploy and maintain self-hosted n8n on VPS via Docker

I care less about demo-style workflows and more about reliability: clear states, safe reruns, clean mappings, and visible logs when something breaks.

If useful, I can share a short breakdown of how I would structure these 6 workflows before starting.

Telegram: @avernick

16

Hi BenKo —

I build exactly this type of system: Gmail API + Google Sheets + Slack + AI classification on self-hosted n8n.

The loop protection and reference number tracking are the two details that matter most here, and I can structure this as a fixed-fee package across all 6 workflows.

Quick questions before I send the final proposal:

  1. Are you using Google Workspace OAuth or IMAP/SMTP?
  2. Do you prefer Claude, OpenAI, or Gemini?
  3. Is your n8n instance already running?

If yes, I can send a fixed-price proposal today and start within 48 hours.

17

Hey updates on this?
Book a call here- https://axonyx.framer.ai/
Talk soon!
Aryan

18

Happy to put a structured proposal together. The pieces that look most interesting: AI email classification across DE/EN/FR (I’d lean Claude or Gemini for multilingual accuracy over OpenAI), and the request status tracking workflow with timestamp logging across all 6 stages — that’s where most teams cut corners and end up with broken funnels. I’d want to lock those down with idempotency and a small audit table early.

Quick questions to size the proposal:

  1. Do you already have a Google Sheets prototype I can review, or is this a clean build?
    1. Is your n8n self-hosted on the team’s infra, or do you need someone to set up the host too?
      1. Any preference on which LLM provider — or should I include cost/accuracy comparison in the proposal?
    3. I’ll send fixed rates per workflow + total package within 24h once we clarify those.
19

Hey @BenKo

I’ve built almost the same setup recently — Gmail API + LLM classification feeding into Sheets tracking with Slack alerts. Happy to build all 6 workflows on n8n (self-hosted) exactly to your spec.

Here’s the approach: • AI labeling with multilingual prompts (DE/EN/FR) using Claude or OpenAI for the 5 categories + 50 client accounts • Gmail API auto-forwarding for hotel updates, keyword-triggered Slack escalation for urgent cases • Sheets tracking with unique reference IDs, full timestamp logs across all 7 status stages • Formula-based dashboard + daily Slack digest with link and summary line • First-contact auto-reply with loop protection (header checks + sender history) • Clean error handling, retries, and a Loom walkthrough per workflow

I’d suggest sticking with Sheets for v1 — easy for the team to inspect — and revisiting Supabase later if volume grows. Fixed price per workflow + total package on request.

Let’s schedule a call to discuss the details further.

20

Hi BenKo,

For this email-classification workflow, the guardrails matter as much as the happy path:

Gmail message → category/urgency/status decision → Sheets update + Slack escalation + human review for anything ambiguous

I would not start by trying to build all six workflows at once. I would prove the classification and error handling on one branch first, especially around urgent booking changes, hotel updates, and client/account matching.

I work async-only and fixed-scope. If you still need a narrow implementation slice, send one redacted sample email per category plus the target Sheets columns/statuses. I can scope whether a first fixed paid sprint is realistic.

I can also show a small public proof shape for the artifact format before you send anything sensitive.

Alex Reed

WorkflowPatch

alex@workflowpatch.com