Urgent.News

What's breaking now, across thousands of outlets.

AI

What Building a Real App with AI Actually Cost: 36 Parts, Three Seasons, About $17

Three seasons, thirty-six parts, one clinic app — and a meter that ran the whole time. Every part of the ClinicLive trilogy ended with a 💸 box, every retro added its season up, and every number came from the build log rather than from memory. This post puts all of it in one place: what the whole thing cost, part by part and model by model, what caught the AI's mistakes, and what the money could…

Building a real app with AI cost approximately $17.00, with the expenses broken down into 36 parts and spanning three seasons. The total cost of $17.00 includes API spend, which is the same for a subscription plan, though it may not appear so on the bill. The running total climbs from $0.40 at the spec for Season One to $17.00 at the ship-it part for Season Three.

The most expensive single reading was $1.30, shared between Parts 1 and 2 of Season Three. The seasons are similar in cost, with the redesign being the cheapest, but the dead ends cost the same as the features and are included in the final number on purpose. Season One, consisting of 14 parts, cost approximately $6.20. Season Two, with 10 parts, cost about $3.60, and Season Three, featuring 12 parts, cost around $7.20.

The breakdown of per-part costs ranges from $0.40 to $0.72, depending on the parts included in the calculation.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

AI Era: Building Web Sites and Configuring Nginx Reverse Proxy Process

Preface As the title suggests, I recently spent a few days building a company official website, and on top of that, I’ve developed several sub‑sites for the company over the past few months.

  • AI streamlines website building process with prototypes, UI design, and coding.
  • Developers work as general-purpose engineers, collaborating on functional modules.
  • Self-testing by engineers before submitting modules to formal testing.

The Local AI Ecosystem Is Quietly Becoming Real

Draft #26 — The Local AI Ecosystem Is Quietly Becoming Real Status: 草稿(积压 #26)| 2026-09-02 | 目标平台: Dev.to / Medium | 联动: 方案 A(本地小模型工作流模板)、gig #2(模型选型评估) Three data points in three days tell a story…

  • Apple Mac Mini and Mac Studio demand drives local AI adoption
  • 104GB model runs on 48GB Mac Mini at 12 tokens/second
  • Local model setup on M4 Pro Mac Mini becomes routine infrastructure

Adding an Image-to-Video Button to a Creator Tool

I'm an independent developer designing a small asset tool for content creators. The workflow starts with generating images and choosing one.

  • Developer adds "Generate video" button to image tool
  • Image-to-video uses GPT Image 2 and Seedance 2.5
  • Resulting video showcases dress movement for fashion content

The 2 AM Silent Failure: What Running AI Agents in Production Taught Me About Stability

Most AI agents don't fail the way they do in demos. They fail later, and quieter: a task runs at 2 AM, fails silently, nobody gets alerted, and you discover it the next morning — a full day of work…

  • AI agents often fail subtly in production, not during demos.
  • MeshCtx's system has 3,728 automated tests across Windows, macOS, and Linux.
  • Layered memory approach with 17 regions prevents forgetting in AI agents.

More from Friday 4 September →