Apuntes sobre construir con IA.

Getting Software to Talk to Each Other

Aligning the Universe

We stand at the threshold of humanity's most profound partnership. Artificial intelligence promises to be our collaborator in solving civilization's greatest challenges—from climate change to disease, from loneliness to…

How We're Making Business Software Talk to Each Other 10x Faster

For decades, integrating data between software systems has been tedious and expensive. The Model Context Protocol changes the equation.

Who Owns AI's Ideas?

Here's a question that's about to matter a lot: when an AI helps you discover something, who gets the credit? This isn't hypothetical anymore. AI systems are writing code, finding patterns in data, even proposing new scientific hypotheses. The old frameworks for attribution—built when tools were dumb—are breaking down.

AI Assembly Line

Stop thinking of AI as a single tool, and start thinking of it as a specialized workforce.

Encapsulating Ideas

Have you ever found yourself explaining the same thing to an AI again and again?

Semantic Vessels

When you hear my words spoken through an AI-generated voice or see me represented by a digital avatar, it might feel strange at first—like something fundamental has been lost. But what’s actually happening is far more interesting. I’m not outsourcing my thinking or sacrificing authenticity; I’m experimenting with a new kind of medium.

Composable Workflows

Using Cloudflare Workers with tools like Airtable and Zapier. When organizations first explore using artificial intelligence, they usually begin by plugging AI directly into their tools. For instance, a marketing team might add an API key straight into their Airtable scripts or frontend interfaces to generate social media captions or product descriptions. On the surface, this seems quick and ef...

The Economics of Toolification

Why More Tools ≠ More Clarity We are living through an economic anomaly. Software creation, once gated by capital and engineering talent, is now virtually free. Thanks to AI code generation and no-code platforms, the marginal cost of developing digital tools has collapsed. What once required funding, a dev team, and months of planning can now be produced in a weekend by a single person with a g...

Not Another Dashboard, Please!

There was a time when building software meant knowing how to code. That era is over. Now, anyone with a decent prompt can spin up a working app in a weekend. No engineering degree. No team. Just an idea, an AI assistant, and enough persistence to keep tweaking the inputs. This should feel like a superpower. And it is—at first.

Rethinking How We Understand People

From Gut Feelings to Contextual Intelligence There was a time when baseball teams chose players the way most companies still build teams today: by gut. Scouts would watch a player swing a bat or throw a pitch and evaluate them based on a set of unspoken heuristics: “He’s got a strong arm,” “He moves like a natural,” “He looks like a winner.” These judgments were built on experience, but they we...

Export Your ChatGPT Chats

Solving the Context Window Problem

Airtable MCP: How I Taught Claude to Talk Directly to My Databases

It started with Anthropic's announcement of the Model Context Protocol (MCP). As someone who loves experimenting with new tech, I was immediately fascinated by the possibilities. Here was a standardized way for AI…

Built an Airtable MCP that lets you chat with your database

Show: Knowledge Management Platform

Agentic AI for Home Services Startup Ayudalocal.com

Simplified Home Services with Smart Automation

Guide: AI-Powered Outbound Outreach on LinkedIn

Human-Guided Automation with AI

Inside Scheduler Tools

The Dynamic Interface

One of the most underrated ways to build something powerful is by focusing on data entry points. Most people think of them as just forms or fields, but they’re much more than that. Done right, data entry points are the foundation of a dynamic, interactive platform—a system where users and employees aren’t just inputting information, but actively engaging with it.

Centralizing Content for Business Intelligence

Integrating AI into customer service

Artificial intelligence is steadily changing the landscape of customer service. But for many organizations, the challenge lies in knowing where to start and how to scale up effectively. Rather than diving in headfirst,…

The fragility of big ideas

“Be very suspicious of a priori arguments on empirical matters (e.g. “AI will certainly kill everyone”), they’re usually wrong no matter how convincing. Reality is endlessly surprising. (Cf. Knightian uncertainty.).) Ray Dalio’s latest reflections on the changing world order showcase a familiar tension: the allure of big frameworks and the limits of their application. His five forces—debt, inte...