惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

博客园_首页
Exploit-DB.com RSS Feed
Exploit-DB.com RSS Feed
P
Proofpoint News Feed
G
Google Developers Blog
B
Blog
Engineering at Meta
Engineering at Meta
阮一峰的网络日志
阮一峰的网络日志
The Register - Security
The Register - Security
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
博客园 - 叶小钗
The Cloudflare Blog
The Hacker News
The Hacker News
D
Darknet – Hacking Tools, Hacker News & Cyber Security
C
CXSECURITY Database RSS Feed - CXSecurity.com
雷峰网
雷峰网
F
Fortinet All Blogs
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
H
Hackread – Cybersecurity News, Data Breaches, AI and More
酷 壳 – CoolShell
酷 壳 – CoolShell
Last Week in AI
Last Week in AI
T
Threat Research - Cisco Blogs
A
About on SuperTechFans
量子位
Recorded Future
Recorded Future
博客园 - 三生石上(FineUI控件)
H
Help Net Security
Help Net Security
Help Net Security
P
Palo Alto Networks Blog
cs.CV updates on arXiv.org
cs.CV updates on arXiv.org
T
Troy Hunt's Blog
W
WeLiveSecurity
V
Vulnerabilities – Threatpost
T
The Exploit Database - CXSecurity.com
Know Your Adversary
Know Your Adversary
Apple Machine Learning Research
Apple Machine Learning Research
Scott Helme
Scott Helme
N
News | PayPal Newsroom
AWS News Blog
AWS News Blog
D
DataBreaches.Net
Blog — PlanetScale
Blog — PlanetScale
MongoDB | Blog
MongoDB | Blog
B
Blog RSS Feed
腾讯CDC
J
Java Code Geeks
Microsoft Azure Blog
Microsoft Azure Blog
TaoSecurity Blog
TaoSecurity Blog
GbyAI
GbyAI
Y
Y Combinator Blog
Hacker News - Newest:
Hacker News - Newest: "LLM"
D
Docker

Jordan Fulghum

I built a private ChatGPT for my family OpenAI has a branding problem Post-primary-care On Mystery I Used Claude Code + Autoresearch to Trace My Family History Back to Jamestown POST-SOFTWARE (March 2026) Your Product Is No Longer the Center of Gravity Remote Work, AI, and the Disappearing Engineer Mastery Fun vs Frontier Fun 2026 is the Year of Self-hosting VocalMaze - A 60-Second Screen for Cognitive Decline Album Cards: Rebuilding the Joy of Music Discovery for My 10-Year-Old I Prototyped an IDE for How We Actually Code Now Vibecoding Took Away the Fun
Cassette — The Simple Computer in the Cloud
Jordan Fulghum · 2025-11-01 · via Jordan Fulghum

by Jordan Fulghum, November 2025

wow, everything ’s computer

Most software doesn’t need hyperscale. It needs a computer that is fast, predictable, and always on. Over the past fifteen years, developers forgot this.

We buried the simple idea of “a machine that runs your code” under layers of managed services, billing knobs, and new words that make shipping feel like procurement.

Serverless horrors, microservice mazes, surprise bills, per-seat pricing. Roles, permissions, usage, queues, accounts, metrics about accounts. Each promised simplicity, but together they delivered a boiled frog - one that now requires yearly certification, hawked by the merchants of complexity.

It’s too bad, because a computer is a magic machine. Cassette returns to this simple and powerful idea. It is a computer in the cloud that is yours. You choose small medium or large and the computer in the sky turns on and shows up in your browser and then you get to work. Specifically, you can:

  • Run multiple web applications
  • Host your team’s CI/CD pipeline
  • Give a home to a long running AI agent
  • Store your entire git repository using GitLab or Gitea
  • Run a database like Postgres
  • Monitor your apps and services with Grafana or ELK
  • Self-host Discord bots, media platforms, game servers, or other fun side projects

Here’s why the timing matters:

  • AI agents now work in the terminal, installing, running, and wiring tools in the shell. The command line - what Cassette provides out of the box - is the best runtime for agents.
  • AI-assisted coding is leading to a 100-1000x fold increase in the total number of apps. These apps require a simple and reliable server, not a Kubernetes cluster.
  • Open-source app deployment platforms such as Kamal, Dokku, and Coolify are nearing feature parity, offering zero-downtime deployments and rivaling well-funded silicon valley app platforms
  • 2025 is the Year of Linux, measured by both vibes and real data (Linux desktop market share has increased 150% since 2023).
  • The cloud exodus is happening. 37signals moved Basecamp and HEY off hyperscale to simple Linux servers, saving almost $1.5M per year, joining Geico and Dropbox.
  • Tech twitter leaders such as Pieter Levels (700k followers) frequently remind us that multiple of his high-traffic apps run on a single, simple virtual private server (VPS)

The economies of scale are now running in reverse. Hyperscale promised cheaper and simpler, but individual machines are now both cheaper and simpler. A single VPS can replace a mess of “managed” components, and the bill (often less than the monthly cost of a streaming service) is knowable in advance.

I remember how fun it felt building for the web between ~2005 to 2010. It felt…direct. It’s when I learned to program, and I think I learned the right way. Inputs were requests, forms, files. Outputs were HTML, emails, rows in a database. Frameworks like Rails encouraged and supported this clarity, while abstracting just enough that the ‘suck’ disappeared. Deployments were fast and straightforward. Monoliths ruled by default and app models fit in one human head. Small teams shipped quickly because the surface area was small and obvious.

That Web 2.0 era produced durable companies. Basecamp and Shopify lived on single machines that could be explained at a whiteboard. Stripe and GitHub also began on a simple, server-side framework (Rails). All of these have grown enormously since, but I think the lesson is about keeping the interface between code and computer short, especially in the early days. The work moves cleanly.

Given the web’s pace of change, that era might feel prehistoric. Yet the explosion of AI-assisted and vibecoded apps shows that we are still in early innings. GPUs may spark the fire via AppGen platforms, but simple servers will always keep it burning.

Cassette logo

Cassette removes layers until what remains is clear and sturdy. A computer to trust that feels like a personal laptop, only always on. If AWS is the mainframe, Cassette is the MacBook - approachable but strong enough for just about anything.

In a world where agents write code, applications multiply in number, but devs are more fatigued than ever, the computer comes back to center stage. Cassette makes sure it’s the right kind of computer: simple, powerful, yours.