Job Title: AI Automation Engineer (n8n & Voice AI Specialist)

Jobs
1

Job Title: AI Automation Engineer (n8n & Voice AI Specialist)

The Role:
We are a high-ticket AI Automation Agency (AAA) specializing in Voice AI (Vapi) and Autonomous Internal Agents (OpenClaw). We are looking for a technical powerhouse to handle the backend architecture of our client builds. You will turn business requirements into bulletproof, scalable workflows.

Technical Stack:

  • n8n: Advanced workflow construction (Error handling, Split-in-batches, Function nodes).

  • Voice AI: Deep experience with Vapi.ai (Custom Tools/Function Calling).

  • LLMs: Prompt engineering for Claude 3.5 Sonnet and GPT-4o.

  • Infrastructure: Webhooks, REST APIs, Supabase/Airtable, and Google Workspace integrations.

  • Optional: Experience with OpenClaw or self-hosted AI setups.

Key Responsibilities:

  • Build and “Stress Test” the n8n logic for $4k - $10k client projects.

  • Configure Vapi Custom Tools to allow voice agents to read/write to calendars and CRMs.

  • Document workflows so they can be easily duplicated for future clients.

  • Perform monthly “Brain Updates” and API maintenance for retained clients.

Compensation:

  • Project-Based: $500 – $1,200 per build (based on complexity).

  • Performance Bonus: % of the setup fee for early delivery/high-quality builds.

  • Growth: Opportunity to move into a Lead Engineer role with a monthly retainer as the agency scales.

How to Apply:
Don’t send a generic resume. Send a JSON export or a screenshot of the most complex n8n workflow you’ve built. If you’ve integrated Vapi with a calendar via n8n, you move to the front of the line.

1 Like
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 [email protected]

Looking forward to hearing from you!

3

Stitching n8n to voice and CRM without logic gaps is where most builds fall apart — not in the tools, but in the handoffs.

Shipped a client a Gmail → HubSpot triage workflow last month: 11 hours/week saved, zero manual follow-up, live in 4 days.

First move today: I’d map your exact trigger points across Gmail/Outlook and HubSpot, then wire a working skeleton you can see running.

Want a free proof-of-concept flow built before you commit?

Richard
[email protected]

4

Hey! I work with n8n from 3 years and also worked with Vapi made outbound calling system for clients in my last job.

Airtable, Google Sheets, OpenAI, webhooks all these I use regularly in my workflows.

Can I send some screenshots of my work? Would love to connect.

5

Hi @Marco_Lucent,

I build and maintain a 100+ node production n8n workflow — AI content pipeline with Supabase state machine, 59 Code nodes, Telegram bot interface, LLM chains (Claude/GPT via OpenRouter), and webhook integrations. It runs 24/7 handling real users.

Core stack:

  • n8n: advanced error handling, batching, complex routing (15 If nodes, 2 Switch nodes)

  • LLM: Claude Sonnet + GPT-4o via OpenRouter, structured JSON output, prompt engineering

  • Supabase: state management, session tracking, vector storage

  • Telegram bots: production bots with inline keyboards, callback routing, multi-step flows

  • Vapi.ai: familiar with voice AI + calendar/CRM integration patterns

Can share a screenshot of the live workflow via DM.

Available for per-project work immediately.

6

Hi Marco,

Sharing screenshots of two complex n8n workflows I’ve built recently.

Workflow 1: Voice automation pipeline using the Vapi API where leads are collected, triggered through Google Sheets, and outbound calls are initiated. Post-call data is processed and passed to an AI agent layer that can qualify leads and trigger actions like HubSpot contact updates and Google Calendar scheduling.

Your paragraph text (9)
Your paragraph text (9)1920Ă—1080 76.7 KB

Workflow 2: An AI agent orchestration workflow triggered via webhook. The agent uses memory and tool-calling to interact with systems like Google Sheets, HubSpot, and Google Calendar to read data, update contacts, and schedule events dynamically.

Your paragraph text (8)
Your paragraph text (8)1920Ă—1080 75.7 KB

Would be great to discuss your current setup and how I’d approach structuring the workflows for your client builds.

Best,
Folafoluwa Stephen

Email: [email protected]
Book a call: Schedule a walkthrough Call session here .

7

Hi Marco,

We built FrontDeskPilot — a production Voice AI receptionist SaaS handling real calls in Ukraine and the US. The stack: LiveKit SIP + ElevenLabs TTS (with custom Ukrainian voice cloning) + Deepgram STT + Claude + n8n orchestration. We target sub-5s end-to-end latency across the full STT→LLM→TTS pipeline.

n8n experience: We maintain 50+ workflow packs covering lead enrichment, outbound campaign orchestration, CRM sync, health checks, and content pipelines. Advanced patterns — error handling, split-in-batches, function nodes, webhook chains, and REST API integrations are daily work.

Voice AI: Direct experience with Vapi (custom tools, function calling, calendar/CRM integration), plus deep work with LiveKit SIP trunking, Twilio, and Zadarma telephony. We’ve built research callers, pitch agents, and mystery shoppers — all orchestrated through n8n.

Broader stack: 12 published MCP servers on npm, full-stack TypeScript (Hono + React), Claude and GPT-4o prompt engineering, PostgreSQL, Supabase, Airtable, Google Workspace integrations, and self-hosted AI setups with Ollama.

We document every workflow for replication and maintain them for retained clients — exactly the model you describe.

Happy to share a demo or discuss scope. Portfolio: flipfactory.it.com

8

I have 3+ years of experience building advanced automation systems using n8n, including complex workflow architecture, API integrations, AI agents, and production-level automation. I’ve worked on implementing AI-driven systems with OpenAI, building structured workflows, and integrating CRMs, calendars, and third-party tools.

I also have experience designing scalable, reusable workflows with proper error handling, logging, and modular structure for client delivery.

I’m also a top supporter of the n8n community and regularly work on building high-performance automation systems.

I’d be happy to connect and discuss how I can contribute to your client builds and delivery process.

9

Hi Marco,

This role matches my daily work perfectly. I build n8n + Voice AI systems in production.

What I’ve built:

  1. AI Voice Reservation Agent (1+ year in production): n8n + ElevenLabs + Google Calendar. Handles restaurant reservations via phone autonomously — error handling, retry logic, monitoring, all production-grade.
  2. WhatsApp AI Sales Agent: n8n + GoHighLevel API V2 + OpenAI GPT-4 + WhatsApp Business API. Automated lead qualification and appointment booking 24/7.
  3. 8-workflow YouTube automation pipeline: n8n orchestrating monitoring, transcription, script generation (GPT-4), video production, publishing, and Shorts generation. All self-hosted.

My n8n expertise:

  • Self-hosted on Docker + Coolify + Traefik (Hetzner VPS)
    • Complex error handling, split-in-batches, merge nodes, sub-workflow orchestration
      • Custom JavaScript in Code nodes
        • Webhook architecture design and debugging
          • MCP (Model Context Protocol) via Claude Code CLI
        • Voice AI: Built with ElevenLabs (same STT > LLM > TTS pattern as Vapi). Production-grade with fallbacks.
      • LLMs: GPT-4o and Claude daily. Prompt engineering is core to my workflow design.
    • Infrastructure: Self-hosted n8n, Docker, REST APIs, webhooks, Google Workspace integrations.
  • Available immediately. Based in Spain, fully remote. Happy to do a trial build.

Hector Gallardo Arcos

https://linkedin.com/in/hectorgallardoarcos

10

Hey @Marco_Lucent

I can handle the backend architecture for your client builds, turning business requirements into scalable, bulletproof workflows that integrate Voice AI and Autonomous Internal Agents.

Here’s the approach:
• Build and stress test n8n workflows for high-ticket client projects
• Configure Vapi Custom Tools for calendar and CRM integration
• Document workflows for easy duplication
• Perform API maintenance and “Brain Updates” for retained clients

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

11

Hey Marco,

I’m Francisco Barrera — LLM Engineer & AI Automation Architect based in LATAM.

This role is exactly what I’ve been building in production for the past 2 years. Here’s my relevant stack:

n8n: I run 4+ production workflows (33+ nodes each) on a self-hosted Docker instance — lead processing, AI chatbot orchestration, outbound call management, and automated scheduling. Full error handling, retry logic, and webhook integrations.

Voice AI: I deployed a production voice agent using both VAPI and Retell AI with Twilio SIP trunking for LATAM phone numbers. The agent handles outbound calling, lead qualification, and books appointments directly into Google Calendar. I migrated from VAPI to Retell AI specifically to cut latency from ~990ms to ~600ms.

Integrations: Supabase/PostgreSQL for lead storage, Twilio (Voice + WhatsApp + SIP), Google Calendar API, Meta Ads webhooks, Gemini/Claude/OpenAI APIs.

LLM Prompt Engineering: My voice agent prompt is at V10, incorporating persuasion psychology (scarcity, reciprocity, autonomy, curiosity gaps). I’ve solved real production issues like LLM date hallucination (runtime injection fix) and voice agent speaking before the lead says hello.

Infrastructure: Full VPS setup — Docker, Nginx, Cloudflare, Let’s Encrypt, systemd, UFW.

GitHub: fbarrerainversiones (Francisco Barrera) · GitHub (3 production repos)
LinkedIn: linkedin.com/in/franciscobarrera-ai

I can send a JSON export or screenshots of my most complex n8n workflows on request. Happy to do a quick call to walk through my builds.

Francisco