Build by Parts
Build by Parts
How to build a complex feature by decomposing it into tasks.
In many cases, we can build a feature by giving a coding agent a tech spec. For complex requirements, this is not advisable, for two reasons.
- We often don’t know what we want. Often, we start by building a walking skeleton1 or a tracer bullet2, learn, and make changes along the way.
- The language model behind the coding agent has a limited context window. Keeping each session small is a best practice.
I have a workflow for building a complex feature which I call “Build by Parts”. This essay describes my workflow.
Stop Using Other People's Skills
There are a number of software engineering skills out there and I’m here to tell you to stop using them.
The three most popular plugins for software engineering skills are:
If you have never used such plugins before, I recommend trying one out. If you used one for any non-trivial amount of time, I’d suggest you uninstall it and recreate your own version.
Creating software, like creating anything, is a deeply personal process. To truly master and own software development, you need to also take ownership of the development process. And if you don’t own the agentic skills you use, how can you claim you have ownership of the process?
The Interactive Design Process
Summary
When a developer can already describe what they want, they simply build it. But often, we don’t know what we want. This essay describes how to use AI to design the thing we want to build through an interview process.
Details
Three tiers of knowing what you want to build
- Clear -> Build. No interview.
- Fuzzy -> Interactive Design (non-trivial, but requires planning)
- Unknown -> Complex project. Separate approach.
Interactive Design
- Invoke. The user asks AI to do interactive design for a job. The instructions to AI (skill) are the model of the interactive design process described here, and the AI guides the user through the process so that the process is applied consistently.
- Dispatch. The AI sizes the job, shows the user where they are, and asks: Is Interactive Design the thing you want?
- Interview. One kind of interview or several, until the job is clear (see Kinds of interviews). How the AI does it is up to the AI.
- Confirm domain vocabulary. The AI and the user agree on what things are called, in the user’s terms. This part is essential.
- Surface consequences. The AI asks high-impact questions and helps the user see around corners. The AI decides whether this step is worth running, and the user can always skip it (“I don’t care about these”).
- Review the result. Summary (less than 4 sentences), details, decisions, not in scope. The user reviews until satisfied.
- Produce a tech spec. The AI suggests where the written artifact goes, and the user approves or edits. The written artifact is called a tech spec, which is described later.
Kinds of interviews
Not all interviews are the same. Not all situations require the same interview approach and developers don’t often know which one they need, and one interview may naturally flow into another.
Nemotron Lightning Model
I played around with the Nemotron Lightning 30b model this week. It is one of the most affordable models on the market, at approximately 25% of the price of OpenAI’s ChatGPT Luna model. At the time of this article, Nemotron Lightning costs just $0.05 per million input tokens and $0.20 per million output tokens.
I tested it for my usecase of proofreading my writings. And, well, it did terribly. I had a coding agent evaluate it against DeepSeek v4p1 Flash (also ~4x the price) as a baseline, and what my AI said about Nemotron
Throwing Away What You Built
I spent much of yesterday pouring my mental energy into designing a better AI-assisted code review process. Then I threw it away.
What I learned is that you shouldn’t build every idea you have. I thought I had found a great way to improve the code review process. When I tried to say exactly what the system would improve, the answer was: not much. It was too complicated for the improvement it offered over an agent skill. I didn’t notice this when I was designing it. I had to step away and stop thinking about it before I could see the full cost of building it.
Reviewing My AI-Generated Code
I started using AI coding agents in March of 2025. I’ve always had personal projects before. But adopting AI agents completely changed the scale of my codebase. I am writing tools, web apps, and automations that I never would’ve had the time or energy to write myself. Today, I have more AI generated code in my codebase than code I wrote myself.
This is what Claude says about the growth of my private repos.
Generative AI 2024 Retrospective
Generative AI 2024 Retrospective
2024 witnessed a parade of increasingly more powerful AI model releases, culminating in OpenAI’s groundbreaking “o3”. While I’m certain these advancements really are significant, I don’t think they have translated to noticeable improvements for the average user.
What people are paying attention to, though, are the really amazing product features that are coming out of Anthropic and Google. I’ve heard a lot about people building software Claude’s Artifacts. And Google’s Deep Research product, released a few days ago, has just completely changed search.
Running Your Own LLM UI
This is a review of Open WebUI, an extensible and user-friendly self-hosted WebUI for LLMs.
I recently decided to run my own UI layer for LLMs, as I have some exciting ideas. For those who are not familiar with chatbots, their architecture basically looks like this:
While I’m pretty good at customizing the middleware, I’m not as skilled at customizing the UI layer.
I tried writing my own UI at first, but I quickly gave up when I realized it was beyond my skills as a frontend developer. The next thing I did was look into open-source solutions. There were a lot of choices, but I narrowed them down to these three:
OpenAI and ChatGPT
I had two weeks off at the end of 2022. Telesign closed its operations on the last week of 2022 to give employees well-deserved time off for a year of hardwork and I took another week off in addition to that. Like many technologists, I became captivated by OpenAI’s release of ChatGPT in 2022, and I spent a lot of the last two weeks exploring what OpenAI has to offer.
ChatGPT is a chatbot developed by OpenAI. It has been widely recognized for its impressive capabilities, such as imitating human writing, transforming plain English into code, and making glaringly stupid mistakes. The underlying technology of ChatGPT is a machine learning model known as GPT-3. This model is designed to predict how a human might continue a previous piece of text. For example, given what I’ve written so far, GPT-3 predicted what the rest of this blog post will look like: