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.

• AutoBuilder
• @github@anaclumos@rsc-swr

An unfortunate dope technology that has been stained by a bad use-case (stealing licensed products)

• https://sciop.net/
• Files are in 256kb chunks.
• At any given time, everyone has different chunks.
• The client will regularly ask for the list of chunks my peers have.
• The client will request the rarest chunk to the server
• The client sends chunks to the top 4 inbound chunk senders
• Re-evaluate every 10 seconds
• Every 30 seconds, send chunks to random peers

Articles

Building a BitTorrent client from the ground up in Go

• This mechanism allows us to verify the integrity of each piece as we go. It makes BitTorrent resistant to accidental corruption or intentional torrent poisoning. Unless an attacker is capable of breaking SHA-1 with a preimage attack, we will get exactly the content we asked for
• Now that we have information about the file and its tracker, let's talk to the tracker to announce our presence as a peer and to retrieve a list of other peers. We need to make a GET request to the announce URL supplied in the .torrent file, with a few query parameters
• can communicate using the BitTorrent protocol
• can understand and respond to our messages
• has the file that we want, or at least knows what we're talking about
• Therefore, it's essential to pipeline our requests to keep up with the constant pressure of some unfulfilled requests. This can increase the throughput of our connection by order of magnitude
• Notably, I left out all the glue code, parsing, unit tests, and the boring parts that build character. View my full implementation if you're interested