Blog
Writing on production AI systems, agent workflows, RAG, model routing, backend architecture, and the operational details that make AI useful beyond demos.
Browse categories Browse tags RSS feed
Recent stories
RAG vs AI Agents: Choose the Smallest System That Works
The practical difference between RAG and an AI agent is control flow.
In ai-engineering, software-architecture, Jul 14, 2026
Agentic AI Application Development Beyond the Demo
Agentic AI application development is backend and workflow engineering with a model inside the decision loop.
In ai-engineering, software-development, Jul 14, 2026
Memory Does Not Make an Agent a Good Reviewer
An AI agent can remember the whole project and still make a bad decision.
In ai-engineering, ai-tools, Jul 09, 2026
An Agent Search API Should Use an Interface You Already Know
Web search is rarely one line of work inside an agent.
In ai-engineering, developer-tools, Jul 08, 2026
Nobody Sees the Release Pipeline Until It Breaks
I spent part of a recent Makakoo OS release working on version numbers and package checksums.
In engineering, open-source, Jul 07, 2026
A File Is Not Context Just Because You Uploaded It
Uploading a PDF to an AI tool feels like the work is finished.
In ai-engineering, developer-tools, Jul 05, 2026
The AI CLI Should Be Replaceable. Your Memory Should Not Be.
I have more AI CLIs installed than any reasonable person needs.
In ai-tools, productivity, Jul 04, 2026
If Every Code Review Starts From Zero, Nothing Is Learning
A reviewer finds a missing timeout. The team fixes it. Two weeks later, another service ships without a timeout. A new review catches that one too.
In ai-engineering, developer-tools, Jul 01, 2026
Do Not Crawl a Website Before You Know What Is in It
The first version of a website ingestion workflow usually has one setting: crawl everything.
In ai-engineering, developer-tools, Jun 30, 2026
Your AI Reviewer Should Ask What It Can Delete
AI can write a lot of code before I have finished deciding what the feature actually needs.
In ai-engineering, developer-tools, Jun 28, 2026
Stop Giving Agents Bigger Context Windows. Give Them a Work Log.
Every serious agent demo eventually hits the same wall.
In ai-engineering, agent-infrastructure, May 15, 2026
The AI Assistant Is the Boring Part. Build the Workflow.
Most people use AI for freelance work in the smallest possible way.
In productivity, ai-tools, May 15, 2026
Your AI Agreed With You. That's the Problem.
There is a pattern in how business professionals use AI tools that almost nobody talks about, probably because once you see it, it’s hard to unsee and it implicates almost every AI-as...
In productivity, ai-tools, Apr 14, 2026
The SOW That Almost Shipped Without an Out-of-Scope Clause
The statement of work had been through three internal reviews. A senior engagement manager read it. A practice lead read it. A partner looked at it for twenty minutes before a client ...
In productivity, consulting, Apr 14, 2026
Research Protocols Are Documents Too
The journal sent the manuscript back with major revisions. Reviewers one and two had independently identified the same problem: the sampling frame didn’t account for the 2022 publicat...
In productivity, research, Apr 14, 2026
How to Run Your Q2 Budget Past Three AI Reviewers at Once
Every FP&A team has a story about the Q-close that almost went sideways. Usually it involves an assumption that was internally consistent but factually wrong — a headcount plan th...
In productivity, finance, Apr 14, 2026
Lope para presupuestos, papers académicos y memos al consejo
Lope es un runner de sprints con un ensemble de validadores multi-CLI. Cuando la gente ve “sprint”, asume “código”. Esa suposición es incorrecta, y es lo que más quiero corregir en el...
In ia-engineering, productividad, Apr 14, 2026
Presentamos Lope: Cualquier CLI de IA implementa. Cualquier CLI de IA valida.
``` ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ ████████████████████████████████████████████████
In ia-engineering, open-source, Apr 14, 2026
Por qué construí Lope: el punto ciego del modelo único del que no podía escapar
Una historia de origen corta, porque varias personas han preguntado de dónde vino lope y por qué me molesté en construir otro sprint runner cuando ya hay suficientes.
In ia-engineering, build-in-public, Apr 14, 2026
El 'Cavernícola Inteligente': Cómo gruñirle a los validadores de IA reduce tokens un 50-65%
Hay un modo en Lope llamado caveman inteligente. Está activo por defecto. Reduce el coste en tokens de cada respuesta validadora entre un 50 y un 65 por ciento sin perder ni un número...
In ia-engineering, optimizacion, Apr 14, 2026
Why I Built Lope: The Single-Model Blindspot I Kept Tripping Over
A short origin story, because a few people have asked where lope came from and why I bothered building another sprint runner when there are already plenty.
In ai-engineering, open-source, Apr 13, 2026
The Intelligent Caveman: How Grunting at AI Validators Cuts Tokens 50-65%
There is a mode in Lope called intelligent caveman. It is on by default. It cuts the token cost of every validator response by 50 to 65 percent without losing a single line number, pa...
In ai-engineering, open-source, Apr 13, 2026
Lope for Marketing Budgets, Research Papers, and Board Memos
Lope is a sprint runner with a multi-CLI validator ensemble. When people see “sprint,” they assume “code.” That assumption is wrong, and it’s the thing I most want to fix in the first...
In ai-engineering, productivity, Apr 13, 2026
Introducing Lope: Any AI CLI Implements. Any AI CLI Validates.
``` ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ ████████████████████████████████████████████████
In ai-engineering, open-source, Apr 13, 2026
Featured
-
🌐 AI Agent Protocols: The Enduring Foundation for Developers in 2025
In Artificial Intelligence, AI Development, Protocols, Frameworks, Future-Proofing, -
🚀 Revolutionizing Medical Imaging with AI: UCLA's SLIViT Model
In Artificial Intelligence, Healthcare Innovation, UCLA, AI Models, Deep Learning, -
Taking Control of AI: Why You Should Build Your Own AI Server
In Generative AI, AI Strategy, Business Innovation, Technology Integration, Digital Transformation, -
Nvidia's Vision: The Future of AI, Robotics, and Digital Humans
In Artificial Intelligence, Robotics, Digital Transformation, Technology, -
Implementing RAG (Retrieval Augmented Generation) Made Easy
In RAG, AI, Customer Service, Technology Integration, Business Innovation,