AboutWorkServicesToolsBlogContact Get in touch →
AI Agent + Growth Lead · Philippines

I build self-running growth engines powered by AI agents.

I'm not another freelancer running your social calendar or bidding your ads. I design the systems underneath, the agents that write, localize, price-check, and publish while I'm asleep.

Four live properties prove the model before it ever reaches a client. AI consulting, MarTech architecture, and growth automation, built the same way I build my own sites: shipped, measured, kept if it works.

Philippines · Global Remote Founder & Operator AI-Native Builder
The Problem

Marketing isn't slow because you're not trying. It's slow because a person is in every step.

Most teams have the pieces already. A content calendar. A CRM. An ad account. Someone writing copy, someone approving it, someone pushing it live.

The problem isn't effort. It's that every step waits on a human, and humans don't scale linearly. Add a market, add a headcount. Add a channel, add a headcount. The system never gets faster, it just gets more expensive.

The result:

  • Content that ships once a week when it should ship daily
  • Localization that stalls at 2–3 markets because each one needs a person
  • Campaigns that work, then quietly stop working because nobody's watching the data closely enough to notice

I hit this wall running my own affiliate sites before I ever pitched it as a service. The fix wasn't hiring more people. It was removing the human from steps that didn't need one.

Why Most AI Adoption Fails

The missing piece isn't more AI tools. It's architecture.

Most companies "doing AI" have a chatbot bolted onto a workflow built for humans. Someone still copies the output into the CMS. Someone still checks it against the brand guide by hand. Someone still exports the CSV and re-uploads it somewhere else.

That's not automation. That's a faster typist.

The teams that actually get leverage design the workflow around the agent from the start (inputs, guardrails, human review gate, output) instead of dropping AI into a process built before AI existed.

Architecture is the difference between a tool you use and a system that runs.

My Approach

I build the system that does the work, not another deliverable that sits in a folder.

Most consultants hand you a strategy doc. I hand you working infrastructure, and I run the same kind on my own properties first, so I know where it breaks before your team finds out.

Before any agent gets built, I map:

  • What's actually repeatable vs. what needs a human's judgment
  • Where the current process leaks time, money, or accuracy
  • What "good" output looks like, so the agent has a target, not a vibe
  • Where the human review gate sits, and what it's checking for

Because an agent without a review gate isn't automation. It's a liability with better grammar. Everything I ship keeps a human checkpoint before anything goes live, that's not a limitation, it's the design.

Currently Building

Six sites. One system, tested six different ways.

I don't sell a system I haven't run myself. Here's what's live right now.

AIPriceRadar.com

AI software review, coupon, and price-comparison platform. Helps people find and save on AI tools instead of guessing off a listicle. In alpha, 25+ affiliate partners onboarded and growing.

AI Tools · Affiliate · Alpha

EuroPriceDrop.com

AI-native deal and promo-code platform for Europe — UK, Germany, France, Spain, Italy, Netherlands. Strong organic traffic and conversion already; scaling to 20–30 markets on a hyperlocal content strategy, not a copy-paste one.

Deals · Multi-Market · Scaling

TipidNation 2.0

The Philippines' deal-discovery platform, running since 2017. Now on a rebuilt stack — AI-powered content generation and agentic workflows replacing what used to be manual.

Affiliate · Rebuild · Since 2017

Chocobot AI (choco.bot)

A publication and community about how people actually use AI to ship, write, design, and run businesses. Less "AI news," more "here's what worked."

Publication · Community

WebGeek.ph

Where I started in 2009 — one of the Philippines' largest tech and developer communities, partnered with Microsoft, GitHub, Accenture, Smart, and Globe. Currently rebuilding it for the AI era.

Community · Since 2009 · Rebuild

BudgetDecoded.com

Budget travel guides decoded: real prices, reviews, itinerary, verified monthly Country-by-country budget math for Asia.

Travel · Cost Guides · Alpha
See the full breakdown →
What I Offer

Services designed around systems that outlive the engagement

AI Workflow Automation

Autonomous agents that run growth workflows, revenue ops, and customer processes without someone babysitting them. I wire them up and keep a human gate before anything ships.

AI-Native Growth Systems

Growth built around agents from day one, multi-market campaigns, content automation, continuous optimization that keeps working between check-ins.

Revenue Automation

AI across revenue operations, from acquisition to expansion, the unglamorous backbone that makes growth repeatable instead of lucky.

Content & Localization

Content pipelines and multi-region localization that don't read like machine translation. Shipped across 80+ markets, know exactly where it breaks.

Lifecycle & CRM

Customer lifecycle and CRM automation that stays personal at scale, built on the same segmentation work run for millions of users at Celsius.

AI Consulting & Implementation

Hands-on help designing and shipping AI-native systems, working infrastructure you keep, not a strategy doc you file away.

View Full Services →
My Process

How we work together

01
Audit
Map the current workflow. Find what's repeatable, what leaks time or accuracy, and what actually needs a human.
02
Design
Architect the agent system: inputs, logic, guardrails, and where the human review gate sits.
03
Build
Ship the working system, not a slide deck. Same tools I use on my own properties: n8n, Claude, Astro, MCP.
04
Review Gate
Nothing goes live unreviewed. Set the checkpoint where a human signs off before output ships.
05
Run & Optimize
Watch the real numbers, not vanity metrics. Refine the system based on what the data says, not what looks good in a deck.
Built for your stage
This process compresses or expands depending on what you actually need. A single agentic build looks different from ongoing MarTech architecture, the steps are the same, the depth isn't.
Why Work With Me

Proof over promises

Most people selling "AI systems" haven't run one past a demo. I run six, every day, and they're the reason my sites earn while I sleep.

Instead of asking "what AI tool should we add," I ask: what step in your workflow doesn't need a human, and what happens if we remove one?

Systems over campaigns, a system runs again next week without me; a campaign doesn't
Automate the work, keep the judgment, repeatable tasks go to AI, anything needing taste stays with a person
Data-driven, intuition picks the direction, data confirms it
AI-native by default, the workflow is built around agents from day one, not retrofitted
Build like an owner, I run my own properties, so I think in flywheels and compounding, not quarters
Garmin watch showing a VO2 max reading of 54, rated Superior
"Data-driven" isn't a slide bullet for me. I check this number the same way I check conversion rate, daily, no exceptions.
Who I Help

Built for teams ready to remove the human bottleneck

I work best with teams that already know manual work is the ceiling, not the standard. Including:

Startups scaling into new markets without scaling headcount
Teams automating customer lifecycle and revenue ops
Agencies that want white-label AI automation under their own brand
Founders running lean who need senior AI/growth work without a full-time hire

If you want someone to just run your ad account, I'm not the right fit. If you want someone who designs the system that runs it, we'll work well together.

About

Hi, I'm John

I'm a generalist operator from the Philippines who bets on a simple idea: the future of work is less about doing, more about designing the thing that does it.

John Arce reading a book outdoors during a break from training
Same reason I read as run, breadth compounds. You don't know which mile or which chapter pays off later.

I've never fit one title, marketer, builder, community organizer, operator. I run six live properties myself, from AI tool reviews to deal discovery across Europe to travel cost guides, and every one of them is a test of how far a system can run without me.

Before this, I led marketing technology at Celsius through its scale from under 40 people to 800+, and roughly 2 million users. Before that, I built one of the Philippines' largest tech communities from zero.

Kindle showing a quote about wealth, held up against a lake and mountain
The highest dividend isn't money, it's the system that gives you the time back. Same reason I build them.
Read My Full Story →
Frequently Asked Questions

Before you reach out

What does an AI agent specialist actually do? +
I design and deploy autonomous AI agents that run growth-marketing workflows end to end, content generation, multi-market localization, lifecycle messaging, and competitive research, so the system executes continuously instead of waiting on manual effort.
How is AI-native marketing different from just using AI tools? +
Using AI tools means bolting a chatbot onto an existing process. AI-native means the architecture is built around autonomous agents from day one, they think, decide, and execute across channels, and the workflow is designed so AI is the engine, not an add-on.
What kind of companies do you work with? +
Series A/B startups scaling into new markets, enterprises automating customer lifecycle and revenue operations, and agencies that need white-label AI automation. The common thread is teams that want compounding systems, not one-off campaigns.
Are you available for fractional or project-based work? +
Yes. I take on project-based AI consulting and advisory, plus fractional roles such as Growth Lead, CMO, or AI-focused marketing leadership. Engagements range from a single agentic build to ongoing MarTech architecture.
Where are you based and do you work remotely? +
I work from the Philippines, remotely, with teams worldwide. The base keeps senior AI and growth-engineering work cost-efficient without trading away quality or timezone overlap.

AI Agent Operator & Growth Lead Based in the Philippines, Working With Teams Worldwide

John Arce is an AI agent operator and growth lead based in the Philippines, working remotely with teams worldwide. He designs and runs autonomous AI agent systems (content pipelines, multi-market localization, and revenue operations) across his own live properties (AIPriceRadar, EuroPriceDrop, TipidNation, Chocobot, WebGeek) before bringing the same architecture to client engagements.

What does an AI agent operator do?

An AI agent operator designs, builds, and runs autonomous systems that execute marketing and growth work end to end, instead of a person doing each step by hand. This includes content generation, localization, lifecycle messaging, and competitive research, with a human review gate before anything ships.

Who does John Arce work with?

Startups scaling into new markets, teams automating customer lifecycle and revenue operations, and agencies looking for white-label AI automation. Engagements range from project-based agentic builds to fractional Growth Lead or AI-focused marketing leadership roles.

Let's build a system that runs your growth instead of you running it.

I take on project-based builds, advisory, and fractional roles, from a single agentic build to ongoing MarTech architecture. If that's the kind of thing you need, send a note. I read every one.

Contact

Let's build something intelligent.

Want a system that runs your growth instead of you running it? I take on project-based builds and advisory, plus fractional roles like Growth Lead or AI-focused leadership. If that's the kind of thing you need, send a note, I read every one.

⌘C to copy