Ramp의 AX (회사를 AI로 물들이는 법)

Why? Multi-tenant environments. First, we need to understand a few differences between environments:

• End-user UI
• Agent Runtime Environment
• LLM Server

So

• When you run Claude Code on your local MacBook, the first two are always local. The third is usually the Claude.ai server.
• When you ssh to a virtual private server (VPS) and install Claude Code there, the first two are your remote server. The third is still the Claude.ai server.
• When you run Claude RC on your virtual private server and code from your iPad using the Claude app, the end-user UI is on your iPad, the agent runtime environment is on your VPS, and the server is still Claude.ai.

Most people physically separate their tenancy, such as Claude Code, from their personal vs. work laptops. So in most cases, it's not a big deal.

But when you need multi-tenancy, it becomes super stressful. For example, say you have two different toolkits:

• personal toolkits (personal Notion, personal Sentry, personal Linear)
• workplace toolkits (company Notion, company Sentry, company Linear)

Most MCP auth states or code harnesses don't support profiles, so you can only log in to one.

So therefore... a natural evolution was to have both:

• a personal VPS with all personal toolkits set up
• a workplace VPS with all workspace toolkits set up

to physically isolate tenancies.

Now we've solved the multiple-profile issue, but the client's problems persist. Now let's get back to the environments:

• End-user UI
• Agent Runtime Environment
• LLM Server

All MCP auth or toolkit auth info should always be saved in the Agent Runtime Environment IMHO. However, a surprising number of harnesses tie them to the LLM server (such as Codex Apps or Claude.ai Plugins) or put them in the end-user UI (Claude Desktop or Codex Desktop).

Now the problem is:

• If the auth data is put on the LLM server, you cannot reuse LLM accounts across tenants
• If the auth data is put on the end-user UI, you cannot use the same app to access multi-tenants.

The only way to reliably isolate different auth information is thus:

• You ssh to a virtual private server (VPS) and run Claude Code there. Never use LLM server plugins.

Then

• End-user UI
• Agent Runtime Environment

are both isolated VPS, and

• LLM Server holds no information on the tenancy

This way, you can provide different toolkits, creating multiple dev environments.

• Do one thing, but exceptionally well
• Should be explainable within 2 sentences
• Often known as "Boring"

프로덕트는 그보다... 원초적이고 단일적이고 수평 확장 가능한 하나의 목표를 바라볼 때 탄생한다.

References

• Ideas are worthless
• SaaS

Original Post Link is now almost 14,000 lines of raw PHP mixed with inline HTML, CSS in <style> and raw JS in <script> tags

I did not use TS, flexbox or frameworks except jQuery

A lot of $.ajax() and float

though

It has 1,872 paying customers making $61,808 per month

Original Post Link

— @levelsio (@levelsio) July 3, 2023

another calendar app that does "boring" scheduling for businesses.

2 founders with no funding.

generates $8000 MRR.

boring?Original Post Link

Original Post Link

— staticmaker (@staticmaker1) July 11, 2023

You won't believe it, but this super boring SaaS generates $7,699 in MRR.

A/B test your titles live on YouTube. Test any number of thumbnails and titles for every video.

👉

Original Post Link

Good luck building your next web3 project 🤡

Original Post Link

— Dimi ⚡ Startups.email (@tarasowski) July 11, 2023

A super boring SaaS with $9,747 in monthly recurring revenue.

It converts your pdf bank statements into csv, xls.

👉

Original Post Link

Good luck building your next web3 project 😂

Original Post Link

— Dimi ⚡ Startups.email (@tarasowski) July 7, 2023

Over the weekend my side project,

Original Post Link, blew up, becoming the #3 top referrer to @vercel (second only to organic Google search and traffic from GitHub)

Original Post Link

— Steph Dietz (@stephdietz) July 7, 2023

좋은 스타트업 아이디어는 보통

• 작은 아이디어 (금융수퍼앱x 친구에게 송금쉬운 앱o)

• 한마디로 설명 가능 (5초 이상 설명해야하면 대부분x)

• 만들면 첫날에 쓸 사람 이름 대기 가능

— 설이원 (@e1e1e1_31) August 6, 2022

Pretty remarkable (in a good way) how little GitHub's core design system has evolved in a decade.

First screenshot taken in 2013; second one in 2023!

Original Post Link

— Justin Duke (@jmduke) July 9, 2023

about to spin up 300 niche sites in the next 90 days using drafthorse ai

need 100,000 pageview /mo for $10,000 mrr

300 sites only need average 333 pageviews per month

asset value $500,000

upfront cost estimate about $2,000

this arbitrage will be gone in 18 - 36 months

Original Post Link

— Cody Schneider (@codyschneiderxx) July 9, 2023

Recap of this week's boring businesses:

1. iPhone fixer:

Original Post Link

2. PDF API:

Original Post Link

3. Personal to-do list app:

Original Post Link

4. Word counter for writers:

Original Post Link

5. PNG image compressor:

Original Post Link

6.…

— staticmaker (@staticmaker1) July 9, 2023

This dead simple web app generates $96K/year

• Subscription pricing
• Solves a useful problem
• $8.3K MRR
• Growth: word of mouth and SEM
• Launched only 1 year ago! Original Post Link

Original Post Link

— Pat Walls (@thepatwalls) August 1, 2023