Agile Made Devin AI Possible | Prime Reacts

Summary

В видео рассматриваются проблемы и вызовы, связанные с использованием методологии Agile и Scrum в разработке программного обеспечения. Автор выражает свое разочарование по поводу того, как Agile и Scrum могут привести к чрезмерной загружаемости, эмоциональному выгоранию и неэффективному управлению проектами. Он также обсуждает потенциальные угрозы, связанные с развитием искусственного интеллекта в области инженерии программного обеспечения и предлагает мысли о том, как индустрия может адаптироваться к этим изменениям, сохраняя при этом качество и долгосрочную перспективу разработки.

Takeaways

• 🤔 Agile и Scrum подходы к разработке программного обеспечения вызывают определенные проблемы и недовольство среди разработчиков.
• 🚧 Гантты и Водопад (Waterfall) методы управления проектами считаются неэффективными в сфере разработки ПО.
• 🛠️ Extreme Programming (XP) был альтернативой Гантту и Водопад, но его правила также вызывают критику.
• 📋 Agile Manifesto привел к созданию новых методологий разработки, но также столкнулся с определенными трудностями и проблемами.
• 🔄 Итерации и Sprints в Agile разработке могут привести к чувству бесконечного цикла и эмоциональному истощению.
• 😣 Скром и его ритуалы/церемонии вызывают негативные эмоции у некоторых разработчиков.
• 💡 Необходимость адаптации методологии разработки для работы с AI и машинным обучением.
• 🔧 Разработчики ПО должны отличать структурную работу от простого исправления ошибок и добавления функций.
• 🌐 Открытый доступ к информации и обсуждениям в Twitter может привести к потере контекста и неправильному пониманию.
• 📈 Важность тестирования и качества кода в процессе разработки программного обеспечения.
• 🎯 Для успешного создания кода необходимо многократно создавать и масштабировать проекты, учиться на ошибках и развиваться.

Q & A

Какое впечатление производит видео на тему Agile и Scrum у Carl'a?

Carl выражает свое разочарование и критику по поводу Agile и Scrum, сравнивая их с методами, которые не соответствуют реальным потребностям и вызывают стресс, эмоциональное истощение и проблемы с качеством работы.

Почему Carl считает, что Agile и Scrum приводят к проблемам в разработке программного обеспечения?

Carl считает, что Agile и Scrum создают иллюзию быстрого прогресса, но на самом деле приводят к повторяющимся циклам без реальных результатов, а также к постоянному переключению направления работы, что затрудняет оценку и управление проектами.

Какие основные проблемы Carl'a с методологией Agile?

Carl'а проблемы с Agile включают отсутствие планирования, неприемлемое количество встреч, низкое внимание к документации, а также невозможность контролировать долгосрочные цели проекта.

Какие методы разработки программного обеспечения Carl считает более эффективными?

Carl предпочитает методы, которые позволяют более долгосрочным и стратегическим подходом к разработке, включая традиционные методы, такие как Waterfall, и подчеркивает важность документации и тестирования.

По мнению Carl'a, какое влияние AI, такой как Devon, может оказать на будущее разработки программного обеспечения?

Carl считает, что AI, такая как Devon, может заменить некоторые аспекты разработки, особенно те, которые связаны с монотонными и повторяющимися задачами. Однако он также подчеркивает, что AI не сможет заменить более сложные и стратегические аспекты разработки, которые требуют человеческого инсайта и опыта.

Какие изменения Carl рекомендует в отношении Agile и Scrum для улучшения разработки программного обеспечения?

Carl предлагает пересмотреть текущие методологии и больше фокусироваться на структурных работах, а также на долгосрочных выгодах для проекта, вместо бесконечного выполнения задач из backlogа.

Как Carl оценивает эффективность AI в отношении отчетности об ошибках?

Carl критикует использование AI для отчетности об ошибках, считая это бесполезным и отвлекающим, так как AI не может эффективно определить и исправить проблемы, которые возникают в процессе разработки.

Почему Carl считает, что создание небольших проектов и их масштабирование является лучшим способом изучения программирования?

Carl утверждает, что создание и масштабирование небольших проектов помогает разработчикам на практике познакомиться с различными аспектами программирования, а также научиться оптимизировать и улучшать свой код, что в конечном итоге приводит к созданию более надежных и долгосрочных продуктов.

Как Carl относится к Agile Manifesto и его ценностям?

Carl критикует Agile Manifesto за то, что оно отвергает многие традиционные ценности инженерии, такие как документация и планирование, и фокусируется на более краткосрочных и менее структурированных методах работы.

Какие советы Carl дает людям, которые хотят стать программистами?

Carl рекомендует людям, которые хотят стать программистами, изучать кодирование через создание небольших проектов, а также стремиться к постоянному улучшению своих навыков и понимания программирования, чтобы создавать качественное и надежное программное обеспечение.

Как Carl видит будущее разработки программного обеспечения в контексте развития AI?

Carl считает, что AI будет играть все большую роль в разработке программного обеспечения, но также подчеркивает, что человеческий опыт и инсайт будут необходимы для создания успешных и долгосрочных продуктов.

Outlines

00:00

😀 Интро и проблема с Agile и Scrum

В этом разделе видео Carl вводит зрителя в тему и объясняет свою неприязнь к Agile и Scrum. Он утверждает, что многие люди не понимают важности и сложности работы программиста, что приводит к неправильному восприятию роли AI в разработке программного обеспечения. Carl также делится своими мыслями о том, что Agile и Scrum были вызваны неудачами в управлении проектами и что многие разработчики, которые работают над AI, имеют титлы программистов, что вызывает у него желание насмехаться.

05:02

😐 История разработки Agile и Scrum

Carl рассказывает историю возникновения Agile и Scrum, начиная с 1990-х годов, когда управление проектами было связано с использованием Гантовых диаграмм. Он объясняет, как Agile превратилось из 'водопада' (Waterfall) в 'агильный' подход, который включает итерации и небольшие релизы. Carl критикует Agile за его неспособность завершения проектов и создание постоянных изменений, что может привести к выгоранию (burnout). Он также упоминает Extreme Programming (XP) и его правила, которые стали основой для Scrum.

10:04

😒 Проблемы с Agile, Scrum и Dev

Carl выражает свое разочарование в Agile, Scrum и Dev, подчеркивая, что эти методологии и инструменты привели к деградации качества работы программистов. Он критикует упрощение процессов и потерю внимания к важным аспектам разработки, таким как тестирование и документация. Carl также высказывает свою точку зрения на то, что Agile и Scrum создали ненужные и избыточные 'ритуалы' и 'церемонии', которые не добавляют ценности и упрощают взаимодействие с клиентами.

15:05

😕 Влияние AI на будущее программирования

Carl обсуждает возможный сценарий, в котором AI, такие как Dev, могут заменить некоторые аспекты работы программиста. Он выражает свою опасность, что AI может привести к снижению качества кода и потере рабочих мест для квалифицированных разработчиков. Carl подчеркивает важность документации и структурированного подхода к программированию, чтобы сохранить профессионализм и эффективность в будущем.

20:05

😤 Как улучшить качество программирования

В заключительной части видео Carl предлагает несколько идей о том, как улучшить качество программирования и адаптироваться к возможным изменениям, вызванным AI. Он призывает к более серьезному отношению к документации, структуре и долгосрочным проектам, а не просто выполнять задачи из бэклога. Carl подчеркивает важность создания крутых проектов и изучения кода, чтобы стать лучшим разработчиком и обеспечить будущую стабильность и успех в карьере.

Keywords

💡Agile

Agile is a project management and product development approach that emphasizes adaptability, collaboration, and customer satisfaction. In the video, the speaker criticizes Agile for leading to repetitive cycles of work without completion, causing emotional burnout among team members. The speaker compares Agile to a waterfall, suggesting that despite its flexibility, it ultimately follows a similar linear path.

💡Scrum

Scrum is a framework for managing and completing complex projects, particularly in software development. It is characterized by 'rituals' or 'ceremonies' such as daily stand-ups, sprint retrospectives, and planning meetings. The speaker in the video expresses a negative view of Scrum, associating it with cult-like practices and ceremonies that add little value to the development process.

💡Software Engineering

Software engineering is the application of engineering principles to software design, development, testing, and maintenance. The video discusses the perception of software engineering as a profession and criticizes the deprioritization of key engineering deliverables like comprehensive documentation in Agile practices.

💡DevOps

DevOps is a set of practices intended to reduce the time between committing a change to a system and the change being placed into production. It aims to improve collaboration and communication between development and operations teams. Although not explicitly mentioned in the video, the speaker's discussion about software development and deployment could be related to DevOps principles.

💡Extreme Programming (XP)

Extreme Programming (XP) is a software development methodology that is part of the Agile movement. It emphasizes rapid feedback, frequent releases, and close collaboration with clients. The video speaker criticizes XP for its extreme approach and its influence on the development of Agile and Scrum practices, which he finds to be problematic.

💡Waterfall

Waterfall is a software development model that follows a sequential approach to project management, where each stage of development must be completed before moving on to the next. The speaker in the video uses the term to describe the linear and inflexible nature of traditional project management methods, and later extends this criticism to Agile and Scrum practices.

💡Burnout

Burnout is a state of emotional, physical, and mental exhaustion caused by excessive and prolonged stress. In the context of the video, the speaker associates burnout with the repetitive and never-ending cycles of Agile project management, where the lack of completion and constant pivoting leads to stress and dissatisfaction among team members.

💡Standup Meetings

Standup meetings are short, daily team gatherings in Agile and Scrum methodologies where team members report on their progress, what they plan to do, and any impediments they've encountered. The speaker criticizes these meetings as unnecessary and time-consuming, suggesting they contribute to a lack of productivity in software development.

💡Manifesto for Agile Software Development

The Manifesto for Agile Software Development is a set of guiding values and principles for software development, emphasizing individuals and interactions, working software, customer collaboration, and responding to change. The speaker in the video criticizes the manifesto for deprioritizing key aspects of software engineering, such as comprehensive documentation, and for promoting a culture that may not lead to long-term project success.

💡Code Documentation

Code documentation refers to the practice of creating written descriptions and explanations for the source code in a project. It helps others, including future developers, to understand the code's purpose, functionality, and structure. The speaker argues that the Agile Manifesto has led to a deprioritization of comprehensive documentation in software development, which is essential for long-term maintainability and success.

Highlights

Discussion on the perception of software engineering and its comparison to other engineering fields, such as civil engineering.

Critique of the Agile methodology and its impact on software development, including the author's personal experiences and opinions.

Comparison of Agile to Waterfall development, with the argument that Agile is just a more flexible version of Waterfall.

Misunderstandings and misuse of Agile principles, leading to project burnout and lack of tangible deliverables.

Introduction to Extreme Programming (XP) and its key differences from traditional Agile practices.

Critique of standup meetings and the author's preference for more efficient forms of communication in software development.

Discussion on the limitations of Twitter for meaningful discussions and its tendency to promote unproductive debates.

The impact of the Agile Manifesto on the software development industry and the author's skepticism towards its effectiveness.

The role of Java in shaping modern software development practices and the concept of programming as an engineering discipline.

The emergence of Scrum and its influence on project management within software development.

Critique of the Scrum framework, particularly its rituals and ceremonies, and their impact on team dynamics and productivity.

The importance of documentation in software engineering and the potential for AI to improve this aspect of development.

The need for a change in software development practices to adapt to the rise of AI and maintain the profession's credibility.

The potential threat of AI to software engineers who focus solely on grinding through backlogs without contributing long-term value.

Advice for aspiring software engineers to learn coding through creating and scaling projects, emphasizing the importance of creating sustainable and maintainable code.

The author's closing thoughts on the current state of software development, the impact of AI, and the importance of foresight and skill in coding.

Transcripts

all right so this has to be the craziest

title I've ever seen in my entire

lifetime so of course we're going to

have to do this by the way for those

that don't know Ai and Dev fear caused

by agile and scrum failures I mean first

off all my homies hate agile okay so I'm

loving this already my name is Carl and

today we're going to talk about Devon

again we're not going to talk about is

Deon a software engineer I did that

video the problem we're going to talk

about today is I didn't see the video on

is Devon a software engineer but I will

say that all the people that work on

Devon I forget the name of the company

that does Devon they all have human

software engineer in their title and

honestly like nothing's ever made me

want to like you know the Batman

meme like I just I just I just want to I

just want to slap them you know what I

mean like you are just sof engineer just

shut up you know what I mean dude it

just it's bad it's real bad you

shouldn't do that you shouldn't do that

okay given that no one would consider

for a second the idea that this asphalt

p leing machine was a civil engineer why

do so many people think so little of

software engineering that they would

believe that okay first off I got I got

in there look at me I got in I got in on

it let's go let's go by the way if you

did watch my video on Devon I I am

personally not impressed by Devon okay

so for those that don't know I don't

think Devon's a software engineer let's

go we got mentioned we got the free

publicity boys we got it Devin was a

software engineer the answer as you

probably saw in the thumbnail

is agile but how we got there that's the

interesting part so the story starts as

most of mine do back in the 1990s back

then software projects were a new thing

people only knew one good way to manage

projects and that was a Gant chart and

this is a Gant chart it doesn't work

great for software by the way I started

programming like late 90s and then was

in college in like mid 2000s so I never

even I've never even heard of a Gant

chart okay this looks like Waterfall by

the way waterfall like can we all be

real for one quick second agile is just

waterfall okay agile is just you're

taking a kayak down a waterfall so you

can try to like you can try to adjust

courses but let's just face it it's all

waterfall okay it's all waterfall in the

end there are many waterfalls because

here's most of agile is that you do the

same [ __ ] for like three months and

because you're agile it means that the

product manager can then pivot you so

you're in this ever repe competing cycle

of never actually finishing anything and

getting emotionally rung out and

eventually leading to dark places like

burnout like that that's like the power

of agile like if I could describe agile

in a couple sents that would be it agile

just means you can throw stuff away

without uh forcing a group Suicide yes

you can do it without Force suicide but

you also get for you get you get these

slow inevitable burn of group Suicide

due to the fact that it's just like man

it just repeats and then all a sudden

you're getting somewhere things are

actually looking like you're going to do

something and then they're just like I

mean we're Angel let's just pivot a

little bit and then bam hate my life I'm

literally getting in trouble right now

because my project has no testable

pieces for QA until the project is done

and they're yelling uh and they're

yelling about agile and

[Laughter]

testing I hate agile I hate agile okay

there are all of these dates and

durations and things and with software

it's really hard to estimate how long

something is going to take 3 months from

now when you don't even know things that

are going to happen between now and then

you can manage projects with this and I

have but it's not a really good fit

agree and in the late 90s we got this

alternative called XP extreme

programming there's something kind of

hilarious about the title of extreme

programming a gentle introduction like I

can't why is anything that is Extreme

why is it why how could you ever have a

gentle introduction into it like you

should never be there should be no

gentle extremely ungentle introduction

into

extremely gentle and it differentiated

itself from the Gant chart stuff which

they retroactively renamed waterfall as

being you know more flexible and it was

certainly more flexible and that was a

good thing here are the rules for

extreme

programming um you may have never heard

of extreme programming but if you look

at some of these rules they'll seem

familiar from what you've actually been

doing lately these iterations and small

releases became Sprints interation

planning became Sprint planning Oh look

The Stand stand up meeting Joy yo I hate

standup meetings I think we should make

our own version and we're going to call

radical programming okay I'm gonna can I

can I can I just take a step back here

I'm going to say something that some of

you are going to have feelings hurt

afterwards in fact I'm going to not only

just do that I'm going to do it on

Twitter maximum maximum casualties I

don't like working with people that like

standups there set it post it let's go

how do you feel about that how do you

feel about that okay you know what I

just said right there you know what I

just said I went in I went in I don't

want to work with people that want more

meetings Okay I want people that are be

like this meeting better be an email or

the most productive thing I've ever had

because we're not wasting time around

here okay I don't want any of that I

don't want none of it context though

dude this is Twitter there's no context

on Twitter okay if you want context and

you want discussion don't go on Twitter

Twitter is use ful for only two things

platitudes that are completely worthless

and Vim versus the world arguments

there's like nothing else okay there's

nothing else and if you think there

should be something else you're

wrong also [ __ ] in the bio all right

now I I didn't think I say [ __ ] the

bile today but here we are here we

are but a lot of stuff didn't last back

then we had release planning and a

release schedule but a concept

PA program to be fair with this kind of

stuff if especially since this was like

early 2000s there are still like the

majority of stuff being created was

being created in like uh C++ land Java

goys right like that they're actually

shipping out hard products they're not

doing the internets you know what I mean

and so like Internets the idea of

release planning in a release schedule

doesn't make a lot of sense you know

what I mean it just doesn't because

there's not like some definite date it's

not like hey we need to drop our product

on people's desktop and we don't have a

chance to update it very easily so we

better get this one thing right you know

like that's a much different feel back

then than with the internet the

internet's like I don't know we might

get it right today and if we don't I

don't

care programming turned out to be not

very popular and we stopped having as

much emphasis on testing as we probably

should by the way code the unit test

first extreme programming birth tdd

those sons of [ __ ] those sons of

[ __ ] were the ones that did this

didn't they use Collective ownership is

that communism mentioned did we just get

communism and tdd tdd is communism is

that what I'm hearing right now nobody

does tdd correct is that is that is that

what I'm hearing right here right now I

think I am I think I actually am

unfortunately for XP in 2001 we got

Windows XP so that branding wasn't going

to stick so not long after we got

Windows XP we got the agile

Manifesto generally you should never I I

why Manifesto whenever I hear the word

manifesto I just

assume whatever I'm about to read is

going to be unhinged can we just all

agree that somehow that word manifesto

at least in my head contains like uni

bomber vibes okay serial killer Vibes

yeah it's just I've never like maybe I

don't understand what a Manifesto is

maybe I have like a wrong definition on

it but it's it it seems terrible almost

every single time I've ever heard it we

gave up on agile when we realized we are

more agile without agile yeah yeah

absolutely absolutely Mortal Kombat nice

um and it came out it was more fuzzy

than the XP rules uh we got these four

values here by the way I've completely

forgot that this is a video about Devon

and I'm just completely triggered about

about Agile development and standups and

all this I'm not even sure how we got

here but however however this is going

to link everything we're talking about

to to Devon I just cannot wait I I am so

I'm super excited I'm super excited I've

completely forgot that this was an AI

video let's go let's go we deprecated

processes documentation contracts and

plans which I guess at the time seemed

like a good way to distance this

methodology from waterfall okay but I

think this planted the SE That Grew into

our current problem and the year after

that we got the scrum Alliance and it

was way less fuzzy I hit narrowed things

down to what it called rituals and

ceremonies that were either four or this

document single-handedly has probably

created more pain in anyone's life than

anything else okay cult cult called just

ceremonies we just have a couple

ceremonies that's all we're doing it's

not a big deal all I need is for you to

bring a little bit a little bit of goats

butt it's not a big deal five of them

depending on whether you count the

Sprint itself as being a a ritual or a

ceremony I don't know exactly why they

did this I don't know exactly what they

were thinking or why this happened what

I do know Java is the answer Java put

the the ultimate spin on programming

because there was a while there where we

thought you could actually engineer to a

craft

programming aava yeah know true it's

true though it there was a time there

when yava first came out that there's

this idea that that like we could

actually take computer science from a

science into a hard like into like a an

engineering

discipline and that there could be like

actual methodology like methodologies to

to building stuff and like to be fair it

was semi true like you can't if

everybody does the same thing and knows

all the patterns you can build software

together and any person is like

replaceable like an engineer but you

also need to edit like 50 files to add a

little bit more

Json to an endpoint so it's like what's

the trade-off here to have completely

replaceable Engineers that have to

program the world's worst spaghetti coat

ever but because there's patterns that

actually kind of make sense when you

really think about it or just get in

there and mess some [ __ ] up right we

have

3,417 scss files that's crazy wait hold

on no please tell me this is is this

true if this is true hold on Jeremy

Jeremy I just did I see that correctly

Jeremy Jeremy can you please tell me if

this is true and if this is true can you

please tell me it's one like class

profile just [ __ ] talking Damn Damn son

that would I would have been so happy I

would have been so happy not true I've

seen it but not currently okay well damn

what a let down Jeremy way to let

everybody down in chat now can we get

back to the AI

chat is that being a project

manager if you can keep the customer

focused on one Sprint at a time and two

weeks at a time yeah and not look at the

big project plan and not look at the

whole calendar it's a lot easier to keep

the customer from asking questions like

weren't we supposed to have been done by

now and how much longer is this going to

take and wow isn't this costing a lot it

makes it a lot easier to stretch jobs

out which I can say I have seen a lot of

unscrupulous development shops take

advantage of that one just this morning

but the relationship between scrum and

Consulting projects deserves its own

video so I'll be worrying about that

later I do want to give a fa a fair one

on that one which is that scrum

naturally tries to reach into Consulting

because Consulting is like a hard

deadline activity and scrum makes this

unfounded promise that you can

actually you know do the thing and

measure it out and make consistent

progress towards a

goal the only problem with honestly the

only problem with scrum is that it

doesn't take into account that people

work on the

product does that mean where're this is

this where we're going so what do you

think about when you think about an

engineer if you're like most people you

think

about engineering drawings and the

important thing about engineering not

being so much the output of the engineer

but the decisions and the thoughts that

went into doing that this man is using

entirely too much circly device in which

I've currently forgot its name

okay and he's he's actually making

circles on a blueprint that already has

a circle I don't I don't know what this

guy's doing okay something about this is

just a a protractor a protractor isn't

it no it's not a compass it's a

protractor isn't it you

ape y'all we need okay that's it we're

having a standup everybody get in line

we're doing a standup and we don't do

that because we don't really do

documentation and I get it because

documentation is boring and nobody wants

to do documentation facts but in this

case an LM might actually help us it's

quite possible we're going to be able to

get the large language models to help us

do a better job of documentation with

less effort doubt but we need to do

something because if we want to be taken

seriously as a profession we need to

grow up this is on the left the example

deliverables from an engineer that I got

off a website you can see the link down

in the description and on the right

we've got the manifesto for agile

software development what you see on the

left if you look at it for a sec is that

a bunch of that stuffff that an engineer

thinks is important explicitly

deprioritized and agile there's you know

just a lot of stuff here that we've

explicitly just deprioritized and made

people think not so much that we don't

do but these are not things that people

associate with with software

projects working software over

comprehensive documentation okay okay

maybe you know maybe the agile Manifesto

wasn't all bad

okay maybe it wasn't completely wrong so

what are we left with what are our

deliverables well we stand in a circle

we take tickets dang it okay hold on

vocabulary a

protractor ah shoot damn I'm way off

okay okay you you're right y'all right

yall right I'm

wrong we bang out some code we check off

the tickets we stand in a circle we

answer the same three question questions

day after day after day we take tickets

yesterday you get the idea every once in

a while we'll have a demo

meeting and then we'll have a calm and

rational Sprint

retrospective then we have a planning

meeting where we try to fit a bunch of

t-shirt sized stuff into a blatantly

arbitrarily sized bucket with the fewest

amount of things hanging out wow what

kind of crazy pieces are these and by

the way if you designed this by accident

and had this as your last piece tell me

this is not just like the most

satisfying thing ever this guy is on to

something here he's cooking okay I

actually think that we're going

somewhere okay we're going

somewhere and then we go back to

standing in a circle again I am so going

to get cancelled oh well so given that

is it really so surprising that so many

people think that the job of a software

engineer can get done by a fancy

autocomp completing stochastic parrot so

what do we do about it well I don't have

a magic wand I do have a bunch of ideas

we'll be talking a lot more about them

in future videos okay but I think that

software development is going to need to

change and the way that we do things is

going to need to change both to take

advantage of

llms but also to survive them but

regardless of what we do it's going to

take us a long time to dig out of this

public perception but it starts with

making a distinction between the

structural

work and banging out and checking in

features and bugs

fixes think about whether or not the

time you're spending has long-term

benefits to the project or if it's just

yet more inexhaustible backlog turn

because Devon is coming for your backlog

and I think that's great because I hate

grinding the backlog if that's all you

do grinding the backlog is your job then

Devon will be coming for you too and

that's not a place you want to be so for

now best of luck to you let's be careful

out

there

damn damn that was good I liked it I'm

liking it I'm hitting that subscribe

check hey go go like that video go like

go like the video that was great that

was fantastic Ain't No Way um okay so

obviously that that last part with like

if all your job is is gretting the

backlog Devon's coming for you in some

sense he's correct in some sense he's

obviously not correct um in the sense

that your backlogs filled with bugs and

various things and and new features and

ideas and all that kind of stuff you

know backlogs are filled with the whole

kitchen sink and so this this a

that Devon can actually do any of that

stuff I don't buy it like honestly I

don't I don't buy most of that in the

sense that I just don't think we are

this iteration of AI I don't think is

sophisticated enough to make meaningful

changes that aren't just a

huge huge pain in the ass 99% of the

time we've already seen how well LMS are

reporting bugs and curl exactly like

we've already seen this and it was awful

it was actually genuinely awful for

those that don't know it there's a video

I forgot what I call it uh it's like our

horrible future and open source and it

was just like just terrible what they're

doing they're reporting they're just

literally turning out development hours

into fake

investigations it's it's a worthless

waste of time and so I I'm not the i in

llm stands for intelligence yeah I I

actually fully agree with that statement

and so I think people just have this

really odd

view that I I this is what I think is is

that the first time you used an llm

right cuz they're they're pretty new and

the first time we did it it was

incredible right like you you saw it you

did it the first time you opened it up

tried it out and it gave you the yeah

like the Bert Reynolds riding a horse on

Dolly and you're just like damn man this

is incredible I can't believe it just

generated that right and you were you

were shook right and then you tried like

some code and you're like yo write me a

python function to to to graph x and y

and then it did it and you're like

damn this is so cool this is going to

take over the world but when you try it

in real in practice like in real

stuff LMS are the same uh as every new

thing when it works it's amazing when it

fails it kills kittens in a fire yeah

yeah this is this is exactly what

happens do you want to kill do you want

to kill kittens in a fire do you I don't

think so anyways uh for all those that

are still worried I know there's a lot

of people that watch my channel I get

questions like every single day like oh

should I even learn computer science oh

my gosh should I even go into software

engineering oh my goodness should any of

these things happen here's my thing to

you uh that code bases okay code bases

survive because there's one engineer

that has the foresight to be able to

write code that's at like the 90th the

the you know the 90th through the 90th

percentile of skill and ability okay

this one engineer puts a bunch of stuff

on Rails uh then quits cuz he's like

peace out I got better I got I got

better money somewhere else leaves and

then a bunch of you know then a bunch of

the people that are remaining just take

this and keep running with it until it

just absolutely goes off the rails and

then things do not work any longer right

like this is just software engineering

101 if you don't think this is the case

wait until you you find out wait till

you do it llms is like being able to

take code that's generated right out of

the you know right out of the median

right and so you're not getting your

software developed by this person and

putting the rails on here you're

actually getting the software and

putting the railing on by this person

which means that your expected Lifetime

on your software is going to be like

this you're GNA be like oh yeah it's

working and boom it's GNA just it's

gonna suck it's just GNA suck whereas

the other one it's still crashes and

burns like this guy didn't have all the

foresight in the world it's going to go

it's going to go and then maybe it

crashes and burns but man that distance

right here this is the difference

between your startup making it and your

startup not making it okay learn to code

take your time become good the best way

to learn how to code is to create crappy

projects that's it it's that easy create

a project scale it until it completely

fails create another project scale it

until it fails you just need to you need

to do that like 15 times because then

you see I can write a code I can write

code such that it will work for a while

you know maybe it works like maybe you

could write code that is right for like

six months if you could write code

that's right for six months that is

incredible like honestly that's

incredible that means you had enough

foresight not to over abstract write the

right interface and it just works like

that's good that means you're that means

you're you're improving if you think you

can write code that will never need to

be Rewritten you're writing quick sort

or you're diluted like that's it sorry

hate to break it to hate to break it to

you the

name is the primagen