Latest posts
-
What Is Harness Engineering? Keeping AI Agents Reliable
Summary — Harness engineering means designing the mechanisms, rules, and techniques that let an AI agent keep pursuing a goal reliably over a long task. No matter how good the underlying model gets, without a harness an agent will end up convincing itself it’s “done” or wrapping up a task too early. The previous article […] -
Prompt Engineering vs. Context Engineering: What Is the Difference?
Summary — Prompt engineering is know-how about “how to phrase what you say” to an AI. Context engineering is know-how about “what information, structured how, to pre-load” before you ever say anything. The former focuses on a single instruction; the latter focuses on the information environment that needs to already be in place before that […] -
What Is an AI Agent? Context, Subagents, Skills, and MCP Explained
Summary — Unlike a chatbot that only answers questions, an AI agent is a program that creates files, runs commands, and carries itself through multiple steps on its own. This article walks through, in order: the Context Window, where an agent holds what it currently knows; the Subagent, a way to split roles across agents; […] -
Setting Up an AI Agent Practice Environment (Part 0: Before You Start)
Summary — The hands-on tutorial in this series is written so you can follow along with any AI coding agent, but it’s much easier to follow when everyone starts with the same tool and sees the same screen. This article covers Anthropic’s Claude Code specifically, from installation through confirming it “actually runs.” Writing code starts […] -
We Connected ChatGPT to Company Data — Why It Disappointed Us
Ever connected ChatGPT to your internal documents and databases, only to find it fell well short of what you expected once you actually used it? This article breaks down why that disappointment happens, and what has to change to move past it, from a practitioner’s point of view. Between 2023 and 2024, a huge number […] -
One Column Called IS_ACTIVE Confused an Entire Company
What IS_ACTIVE Actually Meant Say a customer table has a column called IS_ACTIVE. Just from the name, you’d assume it means “is this customer currently active.” At this particular company, though, it actually meant “is this customer flagged for contract renewal next quarter.” Even a dormant customer with an expired contract and zero activity would […] -
What RAG Can and Cannot Do — When You Actually Need an Ontology
“Isn’t RAG enough? Do we really need an Ontology on top of it?” We get this question a lot. This article tries to answer it honestly. It isn’t here to trash RAG and sell you an Ontology — it’s here to draw the line between what each one can actually do and where it stops […] -
What Does Palantir Actually Sell? Not Stock, an Ontology
One thing up front: this article isn’t about Palantir’s stock price or an investment call. This blog doesn’t cover investing at all. Instead, for readers who landed here by searching “Palantir,” this is a practitioner’s explanation of what the company actually builds and sells. The short version: Palantir is less an “AI company” than a […] -
What Is a Multi-Agent System? Getting Agents to Collaborate
Summary — Up to now, this series has been about getting the most out of a single agent. But once a task grows large enough, one agent runs out of context, or collapses under the weight of juggling different roles (research, execution, verification) inside one process. This article covers the basic structure of a multi-agent […] -
Building an MCP Server: Connect Claude to Your Internal Systems
Summary — In Part 1 we touched on MCP (Model Context Protocol) only briefly, as “the standard interface that lets an agent connect to outside tools and data.” This article goes further and actually builds an MCP server, so that Claude can reach into your internal systems — internal APIs, databases, files. If you’ve already […]