An AI consultant for companies operating in Latin America
Latin American operations don't look like the ones most AI products are designed for. Customers write on WhatsApp, not web forms. Documents arrive as phone photos. Income is often informal, and every country has its own rules. I help companies build AI that works in that reality, from Bogotá.
Two kinds of clients
In both cases the work is the same: find the process where AI pays off, put an agent into production with the people who do the work, and measure what changes.
- US and global companies operating in Latin America that need AI adapted to local channels, language and regulation, without sending a team to figure it out on the ground.
- Latin American companies (fintechs, lenders, insurers, real estate and service businesses) that want AI running in daily operations, with measured impact on the P&L.
What's different about operating in Latin America
- WhatsApp is the main business channel. Leads, documents, follow-ups and collections all happen there, under Meta's rules for templates, messaging windows and per-message pricing.
- Spanish-first, and not one Spanish. Tone, vocabulary and formality change by country, and customers notice when an agent sounds translated.
- Documents are messy: scanned certificates, photos of IDs, bank statements in local formats. Agents need to read them and say exactly what's missing.
- Informal and mixed income is common, which makes credit and underwriting rules harder to encode.
- Data protection, consumer and collection rules vary by country, and they have to be enforced in the system, not left to the model.
How engagements work
Work is mostly remote, in English or Spanish. Bogotá doesn't observe daylight saving time, so it shares working hours with US Eastern or Central time all year. On-site sessions in Bogotá when the work calls for it.
- AI diagnostic: process map, the three use cases with the best return, build vs. buy and a plan with owners.
- First agent in production: on WhatsApp, by voice or reviewing documents, with tests, permissions and human approval from day one.
- Fractional Chief AI Officer: a recurring seat with your leadership team, AI governance and a P&L impact dashboard.
Track record
Today I design and run the AI agents of a VC-backed mortgage platform in Colombia: 5,100+ sessions in production, an operations agent that absorbed the routine layer of a coordinator role (about USD 51K a year) at USD 274–400 a month, a broker copilot that went from zero to production in 18 days, and a mortgage plugin published in the ChatGPT directory. Before that I co-founded and ran DORA, a rental-guarantee fintech with 2,000 tenants and a 0.8% default rate, which was acquired.
I also built a WhatsApp CRM with a voice agent that logged 181 leads in its first week, with the source identified for 78% of them. Three brands use it today.
FAQ
- Do you work with US companies that have operations in Latin America?
- Yes. A common setup is a US headquarters with customers or operations in Colombia, Mexico or elsewhere in the region. I work with the US team in English and with local teams in Spanish.
- Which countries do you cover?
- I'm based in Bogotá and most of my production work is in Colombia, but the patterns (WhatsApp-first service, document review, credit rules) apply across Spanish-speaking Latin America. Country-specific regulation is always reviewed with local counsel.
- Can you build the agents, or only advise?
- Both. I design and run agents in production myself, and I can also lead your internal team or a vendor through the build.
