stanleydrew

Still, my point is that my parking space isn't actually mine, so I can't modify anything in the garage.

Presumably over time shared parking areas will get upgraded with charging infrastructure to keep attracting tenants.

idk why we need MCP servers when LLMs can just connect to the existing API endpoint

Because the LLM can't "just connect" to an existing API endpoint. It can produce input parameters for an API call, but you still need to implement the calling code. Implementing calling code for every API you want to offer the LLM is at minimum very annoying and often error-prone.

MCP provides a consistent calling implementation that only needs to be written once.

Technically it's not really much different from just giving the LLM an OpenAPI spec.

The actual thing that's different is that an OpenAPI spec is meant to be an exhaustive list of every endpoint and every parameter you could ever use. Whereas an MCP server, as a proxy to an API, tends to offer a curated set of tools and might even compose multiple API calls into a single tool.

I have a silly theory that I only half joke about that docker/containers wouldn't've ever taken off as fast as it did if it didn't solve the horrible python dependency hell so well.

I don't think this is a silly theory at all. The only possibly silly part is that containers specifically helped solve this problem just for python. Lots of other software systems built with other languages have "dependency hell."

Is this just junior devs who never learned it

Seems more like it's fallen out of favor with senior devs who have moved to Go/Rust.

What risk do you foresee arising out of perverse incentives in this case?

Also means you're not having to do a bunch of isolation work to make the server-side execution environment safe.

One of the design principles of sqlc is that SQL queries should be static in application code so that you know exactly what SQL is running on your database. It turns out you can get pretty far operating under this constraint, although there are some annoyances.

Riza, Inc. (https://riza.io) | SWEs and DevRel Engineers | Full-time or part-time | San Francisco

We use WASM to provide isolated runtimes for executing untrusted code, mostly generated by LLMs. Our customers do things like extract data from log lines at run time by asking claude-3-5-sonnet to generate a parsing function on-the-fly and then sending it to us for execution.

Things we need help with:

* Our janky account management dashboard (Postgres / Go / React / TypeScript)

* Our hosted and self-hosted runtime service (Rust, WASM)

* Integrations and demos with adjacent frameworks and tools (Python / JavaScript / TypeScript)

* New products

We have seed money, but the whole company is currently just me and Kyle working out of a converted warehouse on Alabama St. We’re second-time founders, so we know the risk we’re asking you to take and we’re prepared to compensate accordingly. Send an email to me at andrew at riza dot io or pop in our Discord (https://discord.gg/4P6PUeJFW5) and say hi.

Why do we have to "get there?" Humans use calculators all the time, so why not have every LLM hooked up to a calculator or code interpreter as a tool to use in these exact situations?

I would argue that most sota models do know that they don't know this, as evidenced by the fact that when you give them a code interpreter as a tool they choose to use it to write a script that counts the number of letters rather than try to come up with an answer on their own.

(A quick demo of this in the langchain docs, using claude-3-haiku: https://python.langchain.com/v0.2/docs/integrations/tools/ri...)

Yes, we are doing this at Riza[0] (via WASM). I'd love to have folks try our downloadable CLI which wraps isolated Python/JS runtimes (also Ruby/PHP but LLMs don't seem to write those very well). Shoot me an email[1] or say hi in Discord[1].

[0] https://riza.io [1]:mailto:andrew@riza.io [2]:https://discord.gg/4P6PUeJFW5

Plug in a code interpreter as a tool and the model will write Python or JavaScript to solve this and get it right 100% of the time. (Full disclosure: I work on a product called Riza that you can use as a code interpreter tool for LLMs)

The fix doesn't require removing code though as far as I can tell. Could just add an if statement.

This completely ignores the power of network effects. Even if all regulation were removed tomorrow I doubt the market would accommodate new entrants.

Yes, the word "only" is causing the confusion.

Presumably the code interpreter environment that you get with the OpenAI Assistants API [0] is identical to this ChatGPT environment?

[0] https://platform.openai.com/docs/assistants/tools/code-inter...

What would this even look like? An 83b election is something I file with the IRS. Are you suggesting a company might have me sign a contract committing me to not file an 83b election?

How would they ever find out if I did file, and why would they care?

No, although it could also be cashless.

Early exercise is purchasing shares before your options vest, making you a shareholder sooner and solving a bunch of tax issues. The company retains the right (basically the obligation) to repurchase any unvested shares should you leave the company before fully vesting.

What was the analog form? Just a really long string you could pull with a bell on the other end?

Besides, thinking that QR codes can replace servers seems to indicate to me that the article author has literally no idea what they're writing about, if they don't even understand the basic function of servers...

It just said QR code menus can speed up service, not replace servers. Seems logical to me that if servers are spending time bringing out food rather than standing around waiting for guests to tell them what they want and writing it down that service will generally be speedier.

I could add revocation but then I'd need a place to persist the CRL. I could persist it back to the repo itself, but I didn't want to add more complexity at the start. It's probably worth considering though, so I've added an issue.

It's not. VCs are investment professionals and are trying to avoid adverse selection just like everyone else. Here's a relevant quote from a recent Matt Levine newsletter:

I think that, if you had only five minutes with a world-class trader, and you asked her “teach me the essentials of trading,” probably she would spend the five minutes on adverse selection. The essential lesson is that, if you are being offered a trade, that probably means it’s a bad trade; your job is to understand that thoroughly so you can figure out the exceptions.

A profitable business has either reached market potential, or isn't spending enough on growth.

Yes, but spending money on growth is probably the number one thing VCs like to invest in. If you happen to be profitable, but also can demonstrate a clear path to growth, then VCs will lean in.

I was a bit confused because this document is a draft and I couldn't find any mention of "approval." But here's a link to the published RFC that obsoletes the original UUID RFC 4122:

https://www.rfc-editor.org/rfc/rfc9562

We think about the Retool case a lot. User-defined database connectors were an initial inspiration to start building Riza.

Grandma is getting scammed over regular phonecalls already anyway.

This is not a good argument. Yes things aren't perfect in many areas of technology, but that doesn't mean we should give up.

The better version of this argument acknowledges that there are tradeoffs: allowing side loading may introduce risk but the risk is low relative to the other risks users face and the benefits outweigh the costs.

Hi Vasek, happy to chat sometime soon. Send us an email or DM on Discord?

We'll check out the e2b code interpreter sdk. Looks interesting!

It's a joke, don't think about it too much.

This is actually really clever. Future employers often don't know how to interpret a role like CEO.