Catch what you miss, before it costs you.

Businesses lose revenue to things that fall through the cracks. I build automated systems with built-in safety checks that catch them, and know when to stop and ask a person.

Scroll

Five systems, five common business problems

Working builds for common business problems, tested on realistic data, with every decision logged and a person in control of edge cases.

Results come from test data on fictional businesses.

Lead Qualification System

Portfolio project

Inbound leads scored on evidence the moment they arrive. The AI reads each lead and quotes its source; fixed rules decide the score and where the lead goes. No lead is judged on a hunch.

  • Zapier
  • Python
  • Airtable
100%of 40 test leads routed correctly, up from 88% on the first run, after tuning on those same leads
Read the case study

Dead Lead Reactivation

Portfolio project

Inactive leads re-engaged with the right message at the right time, based on signals in your own data rather than a generic blast to everyone.

  • Zapier
  • Google Sheets
  • Slack
31 of 44leads screened eligible, each with a visible reason
Read the case study

Contract Review

Portfolio project

Contracts pass a fixed-rules completeness check before any AI sees them. Clauses are compared with a written playbook, and a separate check catches risks that only appear when clauses combine.

  • n8n
  • Claude
  • Airtable
ZeroAI cost on an incomplete contract, which is rejected at the start
Read the case study

Order Reconciliation

Portfolio project

Mismatches between orders, payments and fulfillment caught before they become a chargeback or an angry email. Simple matching rules first; AI only on the truly unclear cases.

  • n8n
  • Python
  • Claude
86%of 50 records resolved without any AI
Read the case study

Voice of Customer

Portfolio project

Scattered feedback across channels turned into themes you can actually act on, with a spike alert whenever the monthly count finds an unusual jump.

  • n8n
  • Claude
  • Airtable
14xpackaging-complaint spike (1 mention to 14), flagged when that month’s feedback was counted
Read the case study

What I Build

I build AI-powered automation that saves your team time, reduces errors and keeps a person in control.

Sales and Lead Management

Your hottest leads reach the right rep first

Leads from your forms and ad platforms are checked, filled in with extra details, scored and routed automatically, with an alert the moment a strong prospect comes in.

Dead leads get a second chance

Inactive leads receive personalized follow-ups on a schedule, and anyone who shows interest goes straight to your team.

Finance and Operations

Mismatches surface before they cost you

Your storefront, payment processor and fulfillment records are matched automatically, so discrepancies are flagged every time the check runs. AI steps in only for the truly unclear cases.

Legal and Compliance

Risky terms get caught before you sign

Contracts are checked against your own playbook. Risky clauses and dangerous combinations are flagged, and key dates and obligations are tracked for you.

Customer Experience

Complaint trends show up as they start

Reviews, surveys and support tickets are analyzed together, with an alert when a complaint theme spikes in the monthly count and a monthly insights report.

Custom Automation Systems

A system shaped around how your team works

Don’t see your process above? I’ll learn how your team works, find where time and accuracy are being lost and design an automation system around your tools and workflows.

Start with a free discovery call
Built with
  • Python
  • n8n
  • Zapier
  • Claude

Five steps from first call to live

Every project is built with safety checks at each stage. Here’s how one runs, from the free call and a scoped proposal to a system you can trust.

1

Map the Risks

I map where work slips through today and where a mistake would cost you the most.

2

Design the Safety Checks

I set the fixed rules, mark where AI is needed and place human sign-off points before building.

3

Build and Break It

I build each part, then test it against your real data and the failure cases I planned for.

4

Launch with Sign-off

I go live with your team approving the steps where a mistake would be costly, then show them how to run it.

5

Check Every Decision

Every result is saved with its evidence, so you can check any decision. With optional support, I tune the system as things change.

What to expect

Timeline

Set after the call

Every project is different, so I agree the timeline with you in the proposal, once I know the scope.

Pricing

One fixed price

Quoted after the free call and agreed in writing before work starts. Ongoing support is optional: a monthly retainer if you want me to maintain the system, or a handover if you’d rather run it yourself.

Best for

Small teams and growing businesses

  • Small teams under 20 people
  • Sales and operations teams
  • E-commerce and DTC brands
  • Agencies and service businesses

Common questions

How much does it cost?

You get one fixed price, quoted after the free call and agreed before any work starts. Optional monthly support is a separate retainer.

How long does a project take?

It depends on the scope. I agree a timeline with you in the proposal, before any work begins.

What do you need from me?

Access to the tools involved, a few real examples to test with and one person who can answer questions and approve the key steps.

Can my team run it after launch?

Yes. I walk your team through it at launch, and you can then maintain it yourselves or add optional monthly support.

What if my process isn’t worth automating?

I’ll tell you on the free call, including when the answer is no. You still leave with ideas you can use.

Do you work with the tools we already use?

Usually, yes. I design around your existing tools, and I build with Python, n8n, Zapier and Claude.

Hi, I’m Marloie Menchavez

AI Automation and Workflow Builder

Open to new projects Based in the Philippines

I build automations that know when to stop and ask a human.

I’m a licensed chemical engineer in the Philippines. For nearly two years, I worked as a Process Safety Engineer at one of the largest petrochemical complexes in the country, running hazard analyses and safety reviews for the process units I supported and documenting the reasoning behind each decision.

That job taught me to plan for failure instead of assuming things will work. No single safeguard is trusted on its own. I learned to ask “what happens when this fails?” early, and to keep a record so every decision can be checked later.

Now I build workflows that take repetitive work off a team’s plate, like lead routing, order reconciliation and contract review, using the same four safeguards:

Safeguard 1

Fixed rules

Rules handle everything predictable.

Safeguard 2

AI, only where needed

AI handles the truly unclear cases.

Safeguard 3

Human approval

A person signs off where a mistake would be costly.

Safeguard 4

Record of every decision

Every result is saved with its evidence, so your team can see why.

Let’s find what your
process is letting slip.

Pick a time that works for you. The call is free, and there’s no obligation to continue afterward.

What you’ll get from the 30 minutes:

  • A clear picture of the bottleneck

    We walk through one process that’s costing you time, leads or money and pinpoint where it breaks down today.

  • An honest answer on automation

    I’ll tell you whether it’s worth automating, including when the answer is no.

  • A high-level direction

    A rough idea of what a system could do here and where a person stays in control. The detailed design comes after, if we work together.

  • A clear next step

    If there’s a fit, I’ll send a scoped proposal. If not, you still leave with ideas you can use.

You’ll receive a short questionnaire when you book, so we can make the most of the 30 minutes.

If you prefer, send an email first:

hello@marloiemenchavez.com Replies within one business day Happy to work across time zones