How I used AI to roll my own blogging platform
Julian Moors
I took the decision to roll my own platform after wanting to experiment with agentic coding.
My experience of using AI has always been with using a fully fledged IDE with an agent being used as nothing more than Intellisense on steroids. At the same time of deciding to use an AI coding tool I also decided to try a new text editor.
After installing Zed I clicked around to get an overall feel of what Zed is capable of and I saw an agent panel that spanned half of the width of the app indicating that Zed is giving AI agents first class support.
I then signed up for Zed Pro to enable the agents selection panel and decided to use Claude Sonnet as this AI model was recommended by Zed. I saw very quickly that Claude Sonnet was eating up a chunk of money and I looked at other AI models that I thought performed just as good, but are cheaper and that's when I tried Gemini 3.5 Flash.
(see below the pricing table from Zed Industries' website)
| Model | Provider | Token Type | Provider Price per 1M tokens | Zed Price per 1M tokens |
|---|---|---|---|---|
| Claude Opus 4.5 | Anthropic | Input | $5.00 | $5.50 |
| Anthropic | Output | $25.00 | $27.50 | |
| Anthropic | Input - Cache Write | $6.25 | $6.875 | |
| Anthropic | Input - Cache Read | $0.50 | $0.55 | |
| Claude Opus 4.6 | Anthropic | Input | $5.00 | $5.50 |
| Anthropic | Output | $25.00 | $27.50 | |
| Anthropic | Input - Cache Write | $6.25 | $6.875 | |
| Anthropic | Input - Cache Read | $0.50 | $0.55 | |
| Claude Opus 4.7 | Anthropic | Input | $5.00 | $5.50 |
| Anthropic | Output | $25.00 | $27.50 | |
| Anthropic | Input - Cache Write | $6.25 | $6.875 | |
| Anthropic | Input - Cache Read | $0.50 | $0.55 | |
| Claude Opus 4.8 | Anthropic | Input | $5.00 | $5.50 |
| Anthropic | Output | $25.00 | $27.50 | |
| Anthropic | Input - Cache Write | $6.25 | $6.875 | |
| Anthropic | Input - Cache Read | $0.50 | $0.55 | |
| Claude Sonnet 4.5 | Anthropic | Input | $3.00 | $3.30 |
| Anthropic | Output | $15.00 | $16.50 | |
| Anthropic | Input - Cache Write | $3.75 | $4.125 | |
| Anthropic | Input - Cache Read | $0.30 | $0.33 | |
| Claude Sonnet 4.6 | Anthropic | Input | $3.00 | $3.30 |
| Anthropic | Output | $15.00 | $16.50 | |
| Anthropic | Input - Cache Write | $3.75 | $4.125 | |
| Anthropic | Input - Cache Read | $0.30 | $0.33 | |
| Claude Haiku 4.5 | Anthropic | Input | $1.00 | $1.10 |
| Anthropic | Output | $5.00 | $5.50 | |
| Anthropic | Input - Cache Write | $1.25 | $1.375 | |
| Anthropic | Input - Cache Read | $0.10 | $0.11 | |
| GPT-5.5 pro | OpenAI | Input | $30.00 | $33.00 |
| OpenAI | Output | $180.00 | $198.00 | |
| GPT-5.5 | OpenAI | Input | $5.00 | $5.50 |
| OpenAI | Output | $30.00 | $33.00 | |
| OpenAI | Cached Input | $0.50 | $0.55 | |
| GPT-5.4 pro | OpenAI | Input | $30.00 | $33.00 |
| OpenAI | Output | $180.00 | $198.00 | |
| GPT-5.4 | OpenAI | Input | $2.50 | $2.75 |
| OpenAI | Output | $15.00 | $16.50 | |
| OpenAI | Cached Input | $0.025 | $0.0275 | |
| GPT-5.3-Codex | OpenAI | Input | $1.75 | $1.925 |
| OpenAI | Output | $14.00 | $15.40 | |
| OpenAI | Cached Input | $0.175 | $0.1925 | |
| GPT-5.2 | OpenAI | Input | $1.75 | $1.925 |
| OpenAI | Output | $14.00 | $15.40 | |
| OpenAI | Cached Input | $0.175 | $0.1925 | |
| GPT-5.2-Codex | OpenAI | Input | $1.75 | $1.925 |
| OpenAI | Output | $14.00 | $15.40 | |
| OpenAI | Cached Input | $0.175 | $0.1925 | |
| GPT-5 mini | OpenAI | Input | $0.25 | $0.275 |
| OpenAI | Output | $2.00 | $2.20 | |
| OpenAI | Cached Input | $0.025 | $0.0275 | |
| GPT-5 nano | OpenAI | Input | $0.05 | $0.055 |
| OpenAI | Output | $0.40 | $0.44 | |
| OpenAI | Cached Input | $0.005 | $0.0055 | |
| Gemini 3.1 Pro | Input | $2.00 | $2.20 | |
| Output | $12.00 | $13.20 | ||
| Gemini 3.5 Flash | Input | $1.50 | $1.65 | |
| Output | $9.00 | $9.90 | ||
| Gemini 3 Flash | Input | $0.50 | $0.55 | |
| Output | $3.00 | $3.30 |
For those who have yet to try out AI agents, I would estimate that Gemini 3.5 Flash costs ~£50 per day for 8 hours of continuous use. Bearing in mind users won't be running AI models continuously due to code refactoring and feature evaluating, but I think ~£50 per day is acceptable for fair use.
I hope that better frontier AI models will make the likes of Gemini 3.5 Flash cheaper in the future and perhaps even offer capped pricing at 4-6 hours of continuous use for £100 per month.
I guess you're wondering what features I planned for, right? Although I prompted Gemini 3.5 Flash with features, it came up with some interesting implementations:
The first one was a Google SEO Optimizer.
The second one was a cookies banner directly generated from its own CORS code.
The third one was a cover image uploader, which has allowed me to re-use already existing images should I have a long running post series, for example writing a weekly news post.
You're probably wondering if I would recommend using agentic AI for coding projects...
Yes, but it comes with a couple of caveats, the output code is around 95% useful, meaning the AI agent will write comments very similar to the way humans would, this makes sense if the AI models have been trained on OSS projects being hosted on places like GitHub, also you need to have a good idea of what features your app would need and this can only come from experience.
To prevent token overburn I would start off by implementing small projects at first and build up to more complicated projects after you feel confident with the tools. 😉
If you would like to support me, please consider buying me a coffee:
Comments (0)
No approved comments yet. Be the first to share your thoughts!
You must be logged in to post comments.