> Meta has to know that millenials and younger are giving up on their platforms, they have endless internal data showing it, right?

Do you have a source for that? I don't think it's true when looking at global Meta numbers across _all_ Meta social platforms (FB+Instagram+Threads) combined.

I don't see it at all.

> Typed I/O for every LLM call. Use Pydantic. Define what goes in and out.

Sure, not related to DSPy though, and completely tablestakes. Also not sure why the whole article assumes the only language in the world is Python.

> Separate prompts from code. Forces you to think about prompts as distinct things.

There's really no reason prompts must live in a file with a .md or .json or .txt extension rather than .py/.ts/.go/.., except if you indeed work at a company that decided it's a good idea to let random people change prod runtime behavior. If someone can think of a scenario where this is actually a good idea, feel free to elighten me. I don't see how it's any more advisable than editing code in prod while it's running.

> Composable units. Every LLM call should be testable, mockable, chainable.

> Abstract model calls. Make swapping GPT-4 for Claude a one-line change.

And LiteLLM or `ai` (Vercel), the actually most used packages, aren't? You're comparing downloads with Langchain, probably the worst package to gain popularity of the last decade. It was just first to market, then after a short while most realized it's horrifically architected, and now it's just coasting on former name recognition while everyone who needs to get shit done uses something lighter like the above two.

> Eval infrastructure early. Day one. How will you know if a change helped?

Sure, to an extent. Outside of programming, most things where LLMs deliver actual value are very nondeterministic with no right answer. That's exactly what they offer. Plenty of which an LLM can't judge the quality of. Having basic evals is useful, but you can quickly run into their development taking more time than it's worth.

But above all.. the comments on this post immediately make clear that the biggest differentiator of DSPy is the prompt optimization. Yet this article doesn't mention that at all? Weird.

No they didn't [0][1]. With this leak they're probably negotiating as we speak, which could be why they've deleted the posts.

[0] https://chainthink.cn/zh-CN/news/113784276696010804

[1] https://pbs.twimg.com/media/HD2Ky9jW4AAAe0Y?format=jpg&name=...

https://noyb.eu/en/project/dpa/dpc-ireland

GDPR is entirely unenforced, it's not worth the paper it's written on, and this is due to lobbying. The situation continues to this day. The DPAs simply throw reports of violations into the trash bin.

It's hilariously transparent - Ireland recently (less than 6 months ago) added a former _Meta lobbyist_ to their DPA board [0].

US Big Tech is now spending a record €151 million per year on lobbying the EU [1], and it's completely implausible to believe they're doing that with 0 RoI. "The number of digital lobbyists has risen from 699 to 890 full-time equivalents (FTEs), surpassing the 720 Members of the European Parliament (MEPs). A total of 437 lobbyists now have continuous access to the European Parliament.
Three meetings per day: Big Tech held an average of three lobbying meetings a day in the first half of 2025, which speaks volumes about their level of access to EU policymakers." It's impossible that this doesn't influence things.

[0] https://noyb.eu/en/former-meta-lobbyist-named-dpc-commission...

[1] https://corporateeurope.org/en/2025/10/revealed-tech-industr...