AI Automation Engineer - Local LLM, n8n, Agent Swarm Architecture

Please login or register as jobseeker to apply for this job.

TYPE OF WORK

Any

SALARY

1500-3000 monthly

HOURS PER WEEK

20

DATE UPDATED

Apr 18, 2026

JOB OVERVIEW

We are a fast-growing AI automation company (Viral Growth Media) building a private, self-hosted AI infrastructure on a Mac Studio M3 Ultra with 512GB unified memory. We need an engineer who can deploy open-source LLMs locally, orchestrate multi-agent systems using n8n, and build the AI brain that powers our entire business - without relying on cloud APIs.

This is NOT a basic automation job. We are building a 6-agent AI system that runs 24/7 on local hardware, handling content generation, sales automation, lead qualification, DM outreach, revenue analytics, and self-monitoring. If you've never heard of Ollama, LangChain, or local inference - this role is not for you.

What You'll Build
- Deploy and manage open-source LLMs (Llama 3, Mistral, Qwen, DeepSeek) on a Mac Studio M3 Ultra (512GB unified memory) using Ollama
- Build and orchestrate 6 AI agents using n8n as the workflow engine:
> SYNDICATE: Content script generation from trending topics
> FRACTAL: Automated upsell sequences on Whop
> CLOSER: Automated lead qualification and DM outreach via Gmail + Whop
> PRISM: Monthly promotional offers triggered by astrological events
> ORACLE: Daily revenue and performance briefing
> SOVEREIGN: Meta-controller that monitors all agents, detects failures, auto-restarts, and escalates
- Migrate existing ---------- + OpenAI API workflows to local n8n + Ollama (without breaking production)
- Scale our AI subscription product (ScrollKeeper AI) from 32 to 2,000+ subscribers with proper queuing, batch processing, and monitoring
- Build the backend for ViralOS - a multi-tenant SaaS platform (Stripe billing, feature gating, user auth, data isolation)
- Create inter-agent communication systems (webhooks, shared state via Supabase/PostgreSQL, priority queuing)
- Build monitoring dashboards and Slack/Telegram alerting for all agent activity
- Document everything with SOPs and Loom walkthroughs

Must-Haves (Non-Negotiable)
- Hands-on experience with n8n workflow automation (not just Zapier/Make - we need someone who has built complex multi-step n8n workflows)
- Experience integrating LLM APIs (OpenAI, Anthropic/Claude, or open-source models) into production workflows
- Understanding of local LLM inference - you know what Ollama is, you've run models locally, you understand quantization and model selection
- Strong backend development skills (Node.js, Python, or PHP)
- Database experience (PostgreSQL, Supabase, MongoDB, or equivalent)
- Experience deploying on Vercel, DigitalOcean, AWS, or equivalent
- Comfortable with Hubstaff time tracking (screenshots every 10 minutes) from day one
- Willing to commit to full-time exclusivity after a trial period - no side clients
- Able to overlap with US Eastern Time for at least 4 hours daily

Strong Pluses
- Experience with Claude Code, LangChain, CrewAI, or other AI agent frameworks
- Experience building multi-agent systems where agents communicate with each other
- GoHighLevel experience (we use GHL as our CRM - your AI systems will write data to GHL)
- Experience with Whop, Stripe API, or subscription billing systems
- Mac/Apple Silicon experience (our hardware is M3 Ultra)
- Experience migrating cloud API workflows to local inference
- SaaS architecture experience (multi-tenancy, feature gating, user provisioning)

Compensation
- $1,500–$3,000/month USD depending on experience
- Part-time (20 hrs/week) to start, transitioning to full-time (40 hrs/week) after trial
- Paid via Wise on the 1st and 15th
- Long-term position - this is the core AI infrastructure for a company targeting multi-million dollar scale

About Us
Viral Growth Media has sold 12,400+ AI-generated products, has 447K Instagram followers (all organic, zero ad spend), and is building toward a multi-million dollar AI platform. Our products include a $33 AI scroll scan, an $88/month AI subscription, $888 one-on-one sessions, and $11K-$55K enterprise AI builds. The AI infrastructure you build will power all of it.

We are not a startup hoping to find product-market fit. We have revenue, we have customers, and we have a proven product. What we need is the engineer who can build the self-hosted AI brain that takes us from here to scale.

How to Apply
Send a message with ALL of the following. Incomplete applications will not be reviewed:

1. Links to your portfolio, GitHub, or examples of n8n/automation workflows you've built
2. A Loom video (under 5 minutes) walking through the most complex automation or AI workflow you've built - show the actual n8n nodes or code, not a slide deck
3. Answer this question: "What is the most complex AI agent or automation system you have built? Describe the tools, the architecture, and what problem it solved."
4. Tell me which open-source LLMs you've worked with and whether you've run any locally
5. Your availability (hours per week and EST overlap)
6. Your expected monthly rate in USD
7. Confirmation that you are comfortable with Hubstaff time tracking

Applications without a Loom video will be skipped. If you mention GoHighLevel in your application, also mention n8n or a similar automation tool - this role is primarily about AI infrastructure, not CRM management.

SKILL REQUIREMENT
Node JS Python AI Automation
VIEW OTHER JOB POSTS FROM:
SHARE THIS POST
facebook linkedin
  BENCHMARKS  
Loading Time: Base Classes  0.0011
Controller Execution Time ( Jobseekers / Job )  0.0433
Total Execution Time  0.0452
  GET DATA  
No GET data exists
  MEMORY USAGE  
1,521,144 bytes
  POST DATA  
No POST data exists
  URI STRING  
jobseekers/job/AI-Automation-Engineer-Local-LLM-n8n-Agent-Swarm-Architecture-1626906
  CLASS/METHOD  
jobseekers/job
  DATABASE:  onlinejobs (Jobseekers:$db)   QUERIES: 13 (0.0354 seconds)  (Hide)
0.0003   SELECT *
                                FROM exrates
                                WHERE rate_name = 'USD-PHP' 
0.0005   SELECT *
FROM `employer_jobs`
WHERE `job_id` = 1626906
 LIMIT 1 
0.0016   SELECT *
FROM `employers`
WHERE `employer_id` = 445276
 LIMIT 1 
0.0003   SELECT COUNT(*) AS `numrows`
FROM `t_thread` `t`
LEFT JOIN `t_thread_misc` `misc` ON `t`.`id` = `misc`.`thread_id`
WHERE `t`.`job_id` = 1626906
AND `misc`.`id` IS NULL 
0.0004   SELECT e.business_name, e.logo, e.website, e.rebill_date, e.date_added member_date, hits, DATEDIFF('2026-04-19',ej.date_added) duration_days, DATEDIFF('2026-04-19',e.rebill_date) duration_rebill, ej.*, e.deactivate FROM employers e, employer_jobs ej WHERE e.employer_id = ej.employer_id AND
                                   ((e.user_level >= '500' AND ej.date_added <= e.rebill_date)
                                   OR e.employer_id = '' OR (ej.date_approved <> '2000-01-01' and DATEDIFF('2026-04-19',ej.date_added) <= 14 ))
                                   AND e.deactivate != 1 AND ej.deleted = 0 AND job_id = '1626906' 
0.0003   SELECT *
FROM `employer_jobs_skills` `ejs`
LEFT JOIN `skills_categories` `sc` ON `ejs`.`skill_id` = `sc`.`id`
WHERE `job_id` = 1626906 
0.0018   UPDATE employer_jobs SET hit_counts = '***Apr-18-2026=545***Apr-19-2026=1' WHERE job_id= '1626906'  
0.0006   UPDATE employer_jobs SET monthly_hits = '***Apr-2026=546' WHERE job_id= '1626906'  
0.0013   SELECT date_sent FROM jobseeker_sent_emails WHERE jobseeker_id = '' AND job_id = '1626906' AND status LIKE 'sent%' ORDER BY id DESC  
0.0003   SELECT *
FROM `employer_jobs_skills` `ejs`
LEFT JOIN `skills_categories` `sc` ON `ejs`.`skill_id` = `sc`.`id`
WHERE `job_id` = 1626906 
0.0269   SELECT COUNT(*) AS `numrows`
FROM `employer_jobs`
WHERE `employer_id` = '445276'
AND `date_added` >= '2022-06-08' 
0.0004   select * from teasers 
0.0007   SELECT * FROM skill_categories WHERE skill_cat_id='' 
  HTTP HEADERS  (Show)
  SESSION DATA  (Show)
  CONFIG VARIABLES  (Show)