Vhea Villanueva

AI & Workflow Automation Specialist

I spent 14 years shipping production Java systems. Now I build the automation layer — n8n, Make, and Zapier systems wired to AI — that runs a business's busywork without someone babysitting it. Every system below ships with documentation, not just a working demo.

View systems ↓ Get in touch →
Vhea Villanueva
ADD YOUR PHOTO
your-photo.jpg
AI AUTOMATION · SOFTWARE ENGINEERING
01 — Technology Stack

Tools I build with. Engineering depth behind them.

AI automation is the focus; software engineering, cloud, and business systems are the foundation.

AI & Automation

n8nMakeZapierAIAI Agents OpenAIClaude↯Webhooks APIREST APIOOAuthSStructured OutputsRAGRAG / Vector SearchSupabase

Business Systems

HubSpotAirtableNotion Google WorkspaceSlack

Cloud & DevOps

AWSDockerGit Oracle CloudGoogle CloudPostman

Software Engineering

JavaPythonDBSQL JavaScriptShellPSPowerShell
Experience
14 yrs
Software engineering
Focus
n8n · Make · Zapier
+ AI chat agents
Based in
Philippines
Async-friendly, PHT
Available on
Upwork · LinkedIn
Open for projects
02 — Background

Engineer first, automator second.

Before automation, I was a Senior Software Engineer for 14 years — Java, MySQL, Oracle, and AWS, across Amdocs Philippines, Smart, and SLI. I've shipped systems that had to work correctly the first time and keep working without someone watching them.

That's the standard I hold automation work to. A lot of "automation" online is a fragile Zapier chain that breaks the moment a field changes. I build like an engineer: error handling, logging, and documentation included — not bolted on after a client asks why something silently failed.

I'm AWS Cloud and AI Practitioner certified, and I work across Make, Zapier, n8n, Notion, HubSpot, Airtable, and Google Workspace — usually with an AI agent layer on top that does more than autocomplete: reading context, calling tools, and returning structured data the rest of the workflow can act on.

14 YRS
Senior Software Engineer
Amdocs PH · Smart · SLI — Java, MySQL, Oracle, AWS
CERT
AWS Cloud Practitioner ↗
Cloud infrastructure fundamentals — view credential
CERT
AWS AI Practitioner ↗
Applied AI/ML foundations — view credential
NOW
AI & Workflow Automation Specialist
Freelance — Upwork & LinkedIn
03 — Selected systems

Systems I've built, not templates I've copied.

Each one below is a real workflow with a trigger, a decision layer, and an output. Open a case study for the full breakdown and a link to the walkthrough video.

SYS / 01
Nimbus AI Support Agent
n8n

A self-hosted n8n agent that reads incoming support requests, classifies intent, and either answers directly or routes to the right human.

→ Cuts first-response time; runs unattended 24/7
SYS / 02
AI Lead Intelligence System
Zapier

Enriches and scores inbound leads automatically, then writes structured intelligence straight into the CRM.

→ Case study + walkthrough available
SYS / 03
Lead Intake Automation
Make

Validates, dedupes, and scores every inbound lead, then routes it by budget tier with an AI-drafted follow-up ready to send.

→ Rebuilt through 4 versions to survive edge cases
SYS / 04
Notion-to-Client Reporting
Make

Pulls live project data out of a client's Notion workspace and turns it into a scheduled, branded status report.

→ Removes a recurring manual reporting task
SYS / 05
AI Support Triage System
Make

Classifies and prioritizes incoming support tickets by urgency and topic before they hit a queue.

→ Structured triage layer on top of an existing helpdesk
SYS / 06
AI Jobs Scraper + Resume Optimizer
n8n

Slack-triggered: validates a search query, pulls live job listings, loops through each one generating a tailored resume and proposal.

→ One Slack message produces ready-to-send applications
SYS / 07
ASMR Video Creator
n8n

Generates a new video concept while checking memory to avoid repeats, renders it via Vertex AI's Veo model, and publishes to Facebook and YouTube automatically.

→ One scheduled trigger, zero manual steps, two platforms
04 — FAQ

Questions clients usually ask first.

How is this different from hiring a general virtual assistant?
Most VAs execute tasks manually inside the tools you already have. I spent 14 years as a software engineer before moving into automation — so instead of doing the repetitive task by hand every time, I build the system that does it, with error handling and logging built in so it keeps working when you're not watching it.
What tools do you actually build with?
Mainly n8n, Make, and Zapier for the automation layer, wired to AI (OpenAI, Claude) for anything that needs judgment rather than just data-moving. On the business-systems side: HubSpot, Notion, Airtable, and Google Workspace. Full stack is listed above.
How long does a typical project take?
Depends entirely on complexity — a single-workflow automation is faster than a multi-system build with AI decision layers. Rather than guess here, I scope an exact timeline on the discovery call once I know what you're actually trying to automate.
Do I need to know exactly what to automate before I reach out?
No. Figuring that out is part of the discovery call — a lot of the value is in identifying which manual process is actually worth automating first, not just building whatever's requested.
What happens after the automation is built?
You get the working system plus documentation — not just a demo that only I understand. Error handling and logging are built in from the start, so issues surface clearly instead of failing silently.
Is this a one-off build, or do you offer ongoing support?
Both, depending on what fits. Some clients want a clean one-off build with full handoff; others prefer ongoing support as their systems grow or new tools get added. We figure out which makes sense on the discovery call.
05 — Contact
Have a process that shouldn't be manual anymore?

I take on a limited number of automation builds at a time so each one gets actual engineering attention. Book a free 30-minute call — no pitch, just a look at what's worth automating first.

Direct Email↗ Upwork Profile↗ LinkedIn Profile↗
What happens next
1Book a free discovery call — 30 minutes, no pitch, just a look at what's worth automating first.
2Get a scoped plan — what gets built, in what order, and a realistic timeline.
3Build, test, handoff — with documentation, not just a working demo.
Book a discovery call
Widget not loading? Book directly here →