When Code is Free, Why is Claude is an Electron app?

Source: Hacker News

If code is free, why aren’t all apps native?

The state of coding agents can be summed up by this fact.

Claude spent $20 k on an agent swarm implementing (kinda) a C‑compiler in Rust, but desktop Claude is an Electron app.

If you’re unfamiliar, Electron is a framework for building desktop applications using web tech—HTML, CSS, and JavaScript. It lets you build one desktop app that runs on Windows, macOS, and Linux, and lets developers reuse existing web‑app code. Many apps you probably use every day are built with Electron: Slack, Discord, VS Code, Teams, Notion, and more (list on Wikipedia).

Downsides of Electron

• Size & resource use – each app ships its own Chromium engine; the minimum size is usually a few hundred megabytes.
• Performance – they can feel laggy or unresponsive.
• OS integration – they don’t integrate as tightly with native OS features.

These issues can be mitigated with clever development and platform‑specific code, but they rarely are. The benefits of a single codebase and rapid cross‑platform delivery often outweigh the drawbacks.

The promise of coding agents

Coding agents excel at cross‑platform, cross‑language implementations when given a well‑defined spec and test suite (spec‑driven development article). In theory, this could replace Electron’s benefits: write one spec and test suite, then use agents to generate native code for each platform, delivering snappy, performant apps from small, focused teams.

Why Electron is still used

Even Anthropic, a leader in AI coding tools, still ships Claude as an Electron desktop app—slow, buggy, and bloated. The reasons include:

• The “last 10 %” problem – agents are great for the first 90 % of development, but handling edge cases, real‑world scenarios, and ongoing maintenance remains hard and tedious, requiring substantial human oversight.

• Anthropic’s Rust‑based C compiler – after a massive effort, the compiler hit the limits of Opus’s abilities. The team noted:

  The resulting compiler has nearly reached the limits of Opus’s abilities. I tried (hard!) to fix several of the above limitations but wasn’t fully successful. New features and bugfixes frequently broke existing functionality.

  The compiler is impressive given the time and effort, but it is largely unusable; the “last mile” of development is still a major obstacle.

• Support surface area – native builds for macOS, Windows, and Linux triple the bug‑fix and support workload, whereas Electron’s single wrapper mitigates many platform‑specific issues.

A robust test suite and spec could enable a fully native Claude desktop app, but the overhead of that final development stretch and the increased maintenance burden remain significant.

Bottom line

For now, Electron still makes sense. Coding agents are amazing, but the last mile of development and the expanded support surface area keep Electron as the pragmatic choice for many teams.