geordee

Printing Software

I often write in analogies… how the new AI wave compares to the dotcom boom or the industrial revolution. There is something to be learned from every transformation, whether it is from the factory floor or the financial markets. When we remove the specifics of the technology advancements, it becomes an economic phenomenon or a supply chain problem.

Looking at the way software is written by AI, I cannot help but compare it to 3D-printing in the manufacturing domain. 3D-printing was supposed to revolutionise manufacturing, but as of today it is still a rounding error in the overall manufacturing sector....

Credentials for AI Agents

When we employ people for a job, we verify their credentials - their educational qualifications, certifications and work experiences. In the world of AI agents, similar to people, AI agents will need credentials. Human beings acquire theirs through the education system. We are trained for fifteen to twenty years, and that training qualifies us for certain jobs. The proof is in the certificates issued by universities and the experience letters from the companies where we worked.

AI is recent, and agents are more recent still. So how do we make sure they hold the credentials to perform critical tasks in...

These are a few of my favorite editors

Three decades, twelve editors. From vi in 1995 to Helix in 2026. The languages, jobs, and habits each one carried. A nostalgic look back, written before editors became viewers.

vi. (1995)

That’s how it started. We had Wipro Unix server with dumb terminals. Writing mostly C programs for hobby. C, Fortran

Turbo C Editor. (1998)

Not a real favorite. It had keyboard shortcuts and mouse support though. It took me through final year project to design concrete structures based on structural analysis outputs, constraints and assumptions. C

Tedit. (1999)

Terminal editor of Tandem computers. T6530...

Job Descriptions for AI Agents

There is a pattern in AI user interfaces, especially for the chat-based conversational interfaces. User asks a question or explains the problem. AI responds with a good-enough answer. User then follows up with more relevant information until an acceptable answer or solution is achieved. This suits the probabilistic nature of AI. This was the most common pattern in retrieval-augmented generation (RAG) architectures.

Contrast this with a human-to-human conversation where the first question will be responded with a counter-question or request for clarification, instead of a good-enough answer. As models became more capable and tokens became cheaper, we have introduced reasoning...

Data Labeling is the New Data Mining

In 2011 I worked on Oracle Spend Classification, a data mining project. Machine learning algorithms classified products and parts into various spend categories for better control over the spend, contracts and even the design. Unlike the previous efforts the machine learning ran autonomously with a great degree of accuracy. But it worked on structured data, the product master, purchase orders, goods receipt and so on.

However, the world is mostly unstructured. The things we see, the things we do, the words we speak, the music we play. Computers are remarkably inefficient at making sense of all that. We have to...

AI and Overproduction

In February I decided to move to the command line. I started assembling the tools to live primarily in command-line mode. One of the tools I missed badly was a good Markdown reader. I searched far and wide. There were not many good options. Glow was the most convincing choice, but I was hitting bugs and limitations. So I decided to build one for myself. Armed with Claude Code, I built one in a couple of weeks - a command-line Markdown viewer with a simple editor and readability metrics. I shared it with some friends and decided to let it...

New Wine in Old Wineskins

There is a recurring debate about AI replacing human activities - in coding, design, shopping, negotiation, and medical diagnosis. What we often overlook is that we are asking an inherently probabilistic system to perform tasks that demand determinism and precision.

My view is that we need to redefine the processes themselves.

Let us consider the process of design, whether it be web, software, interior, or industrial. Traditionally, we begin with requirements and exact measurements, translating them into prototypes, diagrams, or CAD models. By the time a design is visualised, significant effort has already been invested. Every iteration from that point...

Products, Platforms, and Protocols

Product Thinking, Product Engineering, Minimum Viable Products… We build products to solve problems. Ship fast, learn faster. The product was the thing, and everything else existed to serve it.

Then someone noticed that every team was solving the same problems. Pipelines, containers, observability, deployment - the same foundations being laid over and over again. So we abstracted downward, and platform engineering emerged. Build the foundation once, let products plant themselves on top. The platform handles the how so product teams can focus on the what.

It’s a good deal, if you accept the terms. Comply with the platform’s architecture, follow...

Single-use Systems

It’s nearly impossible to discuss technology today without AI entering the conversation. And increasingly, it’s becoming just as difficult to talk about software without acknowledging a fundamental shift in how it comes into being. I’ve been observing this shift for a while, watching the patterns take shape as AI tools grow more capable, more accessible, and more deeply woven into our development practices.

The pattern I keep returning to is one we’ve seen play out before, in an entirely different domain.

Manufacturing efficiencies enable mass production. Mass production enables cost efficiencies. Push this far enough, and products become cheaper to...

Measure Once, Cut Twice

The old proverb tells us to measure twice, cut once. It sounds like practical wisdom: be careful, avoid waste, get it right the first time. This article challenges that wisdom, in building products, software or otherwise.

I recently watched a woodworking video by Foureyes Furniture where Chris Salomone dismantles this wisdom. Even when it comes to measurement, he explains, the real skill isn’t in measuring precisely. It’s in sneaking up on the fit. Making incremental cuts, testing against the actual work, adjusting until the piece sits exactly where it should. And once such a setup is complete, it’s all...