Enrique Esquinas

Writing in mono-spaced font my whole life. First screenplays, now coding as a software engineer for @MarsBased

[Rob Miles @robertskmiles]

The thing is, with exponential-or-faster processes, there are in practice only two possible times you can react: too early, and too late

[Friendly.rb @friendlyrb]

Just in time to answer "what should I watch on my some summer vacation?", all the Friendly.rb videos are up on our YouTube Channel 🙌

https://t.co/qgfWuVQSGp

[Nate Berkopec @nateberkopec]

You would think when everyone has a couple of MCPs and frontier LLMs to-hand, consulting businesses like mine would be toast. Why pay Speedshop four figures a month to fix your app when "an LLM can just do it for me?"

I've been running at 100% capacity for the past year despite all the advances on the model front. Why?

1. You've gotten more effective, but so have I. What _I_ can accomplish with an hour + $50 of tokens is still 10x what _you_ can accomplish with that.
2. You still don't know what questions to ask. Consultants are _still_ more effective than LLMs at answering the unknown unknowns, whereas LLMs still require you to prompt them. The "analyze my entire codebase" skill you have is nowhere near as effective as what I can do.
3. Trust and blame-ability. Everyone is sitting on piles of now _hundreds_ of pull requests which are generated by agents but they are too nervous to merge without human review. I am that human review and accountability now, or I am the expert who helps you build the software factory which can automate this trust and verification process (which you cannot build yourself because you don't know how to verify software for performance).
4. Not all problems are LLM-shaped. Certain problems are a great fit for "token generator in a loop": fix all my N+1s, make my test suite faster. But others are very, very poorly shaped for LLMs: refactor my codebase to fit Russian Doll caching (it will generate an utter hackjob), prioritize 100 different possible things I could be doing, define the goals of performance work, etc. Drudgery is now automatable, yes, but not all the work we do is drudgery.
5. Not all useful knowledge is in the training data. I have ten years of experiences and knowledge from working on literally hundreds of applications which never made it to my blog or books and does not appear in the training set. I know things the LLMs do not.

There are probably more reasons. And I'm not just sitting on my laurels either, I still feel the pressure to get better every day as the frontier capabilities continue to grow. But the business is still good.

[Enrique Esquinas @esquinas]

Markdown has been and will always be absolutely awesome.

[Enrique Esquinas @esquinas]

@watashi1986 @juanmacias La automatización al 100% es un mito, no existe. Ni siquiera los robots o naves de la NASA que están diseñados por la gente más brillante para no requerir mantenimiento. Al final siempre requieren actualización remota de software y todo tipo de reajustes.

[Enrique Esquinas @esquinas]

@watashi1986 @juanmacias Y además diría que ese avance provocó que muchos más tipos de actividades tuvieran mejor contabilidad, como autónomos, asociaciones, pequeñas empresas, incluso las finanzas personales o del hogar serían muy poco prácticas sin Excel.
No es ni mal ejemplo para contra-argumentar.

[Enrique Esquinas @esquinas]

@GregMolnar Same

[Enrique Esquinas @esquinas]

AI is dominating the "good enough for what it's worth" realm. This is both impressive and disappointing. We humans, we better specialise in fine-tuning our best criteria, and in achieving authenticity and excellence. Any task is more fun this way!
#HappyHolidays

[Enrique Esquinas @esquinas]

@pinzonjulian The framework doesn't force you to use default_scope or accepts_nested_attributes in any way. In my opinion, they are great tools for adding features fast in a quick & dirty way. Most of the time, it's the opposite of what you want, but when you do need it, it's a lifesaver.

[Hiroshi SHIBATA @hsbt]

Finally, https://t.co/cxUnjbtMpj has a new design.

[Hans Schnedlitz @hschnedlitz]

TIL about String.shellsplit. Splits your string using the same splitting rules as various UNIX shells.

#ruby

[Enrique Esquinas @esquinas]

@hschnedlitz This depends a lot on the use case, that's why it'd be a bad idea to put this in String. Also, it's important to not mutate the original arguments by mistake, avoid String#concat, prepend, []...
#Ruby

[RubyCademy @RubyCademy]

When `ActiveRecord::RecordInvalid` gets raised, it actually keeps the record inside the exception. You can grab it with `exception.record`.

Had to use it recently and thought it was pretty cool! 😊

#rubyonrails

[Enrique Esquinas @esquinas]

@strzibnyj @37signals Blackcandy https://t.co/QqOEcH4Ng9

[RubyCademy @RubyCademy]

Rails 8 introduces and recommends `params#expect` over `params#require`, offering more secure and explicit parameter handling! 🤩

#rubyonrails #rails8

[Enrique Esquinas @esquinas]

@fxn Looks exactly like one of those "liminal spaces" shown in @SpaceLiminalBot

[Enrique Esquinas @esquinas]

@igor_alexandrov Actually, the exact result of the code in the card is the empty string followed by a new line character being printed out to standard output. The class method call returns nil alright, but gets implicitly converted to string by puts, which also adds the \n (unlike Kernel#print).

[Enrique Esquinas @esquinas]

@_swanson Also, this doesn't assume 1 user can only have a single account. Or 1 account only belongs to a single user. Both are a common assumption of apps in a early stage.

[Enrique Esquinas @esquinas]

@RubyCademy 😮

[Enrique Esquinas @esquinas]

@dodeja @Dagnan @terminal49 I'm sure slowness comes from yours joins.I suppose you've already tried indexing some stuff to make them faster. So, if some of those joins are used over and over again in different queries/endpoints, you should consider materializing or even persisting intermediate tables.