Blog
Technical deep-dives, product case studies, and what I learn building AI that ships to production.
Guardrails: how to stop an AI agent from doing something stupid in production
An LLM hallucinates, obeys whoever manipulates it, and gets things wrong with confidence. The practical guide to the guardrails that separate a safe agent from an accident waiting to happen.
GuideMCP (Model Context Protocol): what it is and why it matters for AI agents
The no-nonsense guide to the Model Context Protocol: what it solves, when to use it, and why it became the 'USB-C' that connects AI agents to your business tools.
Case studyNine founders, three days, nine projects live
I mentored at Paris Group's Founders AI immersion in Chapecó, Brazil. I walked nine business owners from a real bottleneck to a working product — and presented ConectaAI with a live phone call answered by AI. What makes an immersion actually ship.
GuideAI agent vs. chatbot: the difference that changes everything
A chatbot responds. An agent acts. Understand the real difference between a traditional chatbot and an AI agent — memory, tools, autonomy — and why it decides whether your project delivers results.
Deep-diveHow to ship an AI agent on WhatsApp to production (WABA in practice)
The no-nonsense guide from someone who's done it: from the WhatsApp Business API to the 24-hour window, HSM templates, LLM fallback, and what separates a demo bot from an agent that survives production.
Deep-diveThe architecture of an AI call center for internet providers
How to build voice and WhatsApp support that talks to the customer 24/7: SIP telephony, real-time STT/TTS, LLM with fallback, and integration with the ISP's ERP.
Case studyDavi: a rehearsal and performance studio in the cloud
The case study of an audio SaaS: how to take music rehearsal out of its physical constraints and move it to the cloud — and what that teaches about building products where AI lives at the core.
Deep-diveLLM fallback: architecture for an AI that can't go down
Relying on a single LLM provider means having the availability of the weakest link. How to design cascading fallback for an AI that's resilient in production — without the customer noticing.
Deep-diveFunction calling: how to make an LLM act, not just answer
Function calling (tool use) is what turns a language model from a talker into an agent that executes. How it works, how to design good tools, and the mistakes that break in production.
GuideWABA, BSP, and Meta Partner: the definitive guide to the WhatsApp Business API
What the WhatsApp Business API is, how the 24h window works, templates, BSPs, quality rating, and pricing — the complete glossary for anyone who's going to automate WhatsApp for real.
GuideAI for internet service providers (ISPs): the practical applications guide
Where AI actually delivers results at an internet provider: 24/7 support, duplicate invoices, connection diagnostics, billing, and retention. The guide from someone who builds this for ISPs.
Case studyNexchat: an omnichannel support hub with AI at its core
The case study of an omnichannel SaaS: why support scattered across channels breaks companies — and how to unify WhatsApp, web, and more into a single hub with an AI agent.
GuideWhat is conversational AI? The guide for businesses
Conversational AI explained without jargon: what it is, how it works, where it applies (WhatsApp, voice, customer service) and how to tell if it solves your business problem.
OpinionA product that's live beats a beautiful slide
Why most AI projects die in PowerPoint — and the method I use to take AI from the business problem to production deploy, generating real revenue.
GuideHow much does a WhatsApp AI agent cost? The real math
No magic price: the real cost structure of a WhatsApp AI agent — WABA per conversation, LLM tokens, infrastructure and build. How to budget for real before you hire anyone.
Deep-diveRAG in practice: when to use it (and when not to)
RAG (Retrieval-Augmented Generation) explained without the hype: what it is, how it works, when it actually solves your problem and when it's over-engineering. The guide from someone who runs it in production.
Case studySophia: an AI financial advisor that lives on WhatsApp
The case study of a fintech AI product: how to turn financial education and organization into an agent that talks on WhatsApp, right where people already are.
Deep-diveThe voice stack for AI agents: STT, LLM, TTS and the orchestration
The pieces that let an AI answer a phone call: SIP telephony, real-time transcription, LLM, voice synthesis, and the orchestration that ties it all together with low latency.