Yeah, that one sentence was weird. The rest of it was, I thought, interesting. The idea that genAI is so much better in the attack scenario than in the defense scenario. The fact that Claude has been weaponized. That vendors are, as always, chasing buzzwords. Seemed useful to me.

I have a friend who works at Mastra[0].

Got part way through their tutorial, seemed okay. Haven't used it in prod, though.

0: https://mastra.ai/

Yes, this is so powerful.

One of my favorite moves is to ask a question that I feel has an obvious answer and then say "what am I missing?" Sometimes I am right, other times I am missing something.

Either way I'm modelling:

- that it's okay to ask questions to which the answer seems obvious

- that it is totally fine not to know everything

Mine is "I'm going to ask the stupid obvious question here", then ask the question.

Some good insights here.

> I'm starting to wonder if we'll need to track prompts with commits, like we track commit messages today. Or design systems assuming the debugger never wrote any of the code.

As we start to truly build software systems with a foundation of genAI, there is no question in my mind that the SDLC will change. I'm not sure how you do either of these suggestions, but they seem like a good start.

> people don’t want to admit that it is a way to senior-free future

Can you tell me more? Everything I've read indicates it affects juniors/new devs more. Is that what you mean by a 'senior-free' future? One in which there are no seniors in 10-20 years because there are no juniors now?

Or something else?

You have answered your question.

If we don’t accept/hire juniors (as companies think they are inefficient) to their first positions then how can they become seniors in their branch?

TBH I don’t have a solution for that.

Their stock price recently passed the split adjusted highs of 2000.

Wonder if we'll see a similar headline about NVDA in 2050.

Thanks. It was 6 characters too long if I added that.

I've been involved with a few MCP servers. MCP seems like an API designed specifically for LLMs/AIs to interact with.

Agree that tool calling is the primary use case.

Because of context window limits, a 1:1 mapping of REST API endpoint to MCP tool endpoint is usually the wrong approach. Even though LLMs/agents are very good at figuring out the right API call to make.

So you can build on top of APIs or other business logic to present a higher level workflow.

But many of the same concerns apply to MCP servers as they did to REST APIs, which is why we're seeing an explosion of gateways and other management software for MCP servers.

I don't think it is a fad, as it is gaining traction and I don't see what replaces it for a very real use case: tool calling by agents/LLMs.

> MCP seems like an API designed specifically for LLMs/AIs to interact with

I guess I'm confused now, I thought that what it explicitly is.

It's the choice from manna[0] all over again.

0: https://marshallbrain.com/manna1

thank you for that read

Yeah, the first time I read it, it was a real eye-opener.

The power of science fiction!

I have a friend who calls LinkedIn "a rolodex that other people keep up to date".

There is some value in posting on LinkedIn, but the real value is that you can go back and find people who are weak connections when you are looking to hire, purchase services, or ask favors.

I think everyone should join LinkedIn and connect to every one of their colleagues that they would work with again. Then, once in a while, keep that connection alive by sending a message or commenting on a post.

It's a long game, but will pay dividends should you ever need to chat with them.