The Future Belongs To People Who Do Things: The 9 month recap on AI in industry

Okay, welcome back. So, we got a bit of a tour to force us to finish and then we can go and grab a drink in the Sydney Sunshine, the early spring Sydney

Sunshine uh and talk about today's uh various activities. So, to finish up, we

various activities. So, to finish up, we have Jeff Huntley. Jeff I got to know, not unlike a couple of other speakers because he started publishing a series

of pieces earlier in the year. And I I as I mentioned I kind of amplify everything I find that things of interest. And then I noticed that he

interest. And then I noticed that he happened to be in Australia.

So we got in touch. We had a chat and I've got Jeff to come and do variations of this talk at a couple of conferences this year and he's done it in lots of

other places uh as well. Uh I love uh I know it's in this one but your particular the future belongs to people who do things and I think that really

does characterize a lot of what these technologies are enabling. It's kind of you know that really is about people and it doesn't necessarily mean software developers and engineers but it's really

about doing stuff. So to close out today uh please welcome Jeff Huntley.

[Applause] Hey everyone. Um, thanks for joining us

Hey everyone. Um, thanks for joining us here today. Um, thank you to all the

here today. Um, thank you to all the speakers who spoke today. It's never

easy doing talks. Um, it gets easier, but it's always hard. If you haven't done a talk before, I recommend that you actually do do a talk. Go to your local

meetup, start putting submissions into a conference. It's some of the best

conference. It's some of the best personal development and learning you can get. It's through the act of

can get. It's through the act of teaching that you uh get refinement of knowledge. So this is the uh future

knowledge. So this is the uh future belongs to people who do things.

And this is really a story. This is

really just a story. I'd like to say that we're in a oh [ __ ] moment of time.

There will be swearing in this talk. Um

and I wrote this blog post nine months ago. nine months ago and essentially what happened was a

engineering director approached me and all the other tech leads over Boxing Day. It says, "Hey, look, I know it's,

Day. It says, "Hey, look, I know it's, you know, it's it's PTO. It's Boxing

Day. Could you like go deep with AI?

Like this is Boxing Day and like curses just come out. ADA is starting to cook.

Goose is starting to cook. Tool calling

is becoming a thing. It's like, no, I actually need the tech leads to go play with AI o over the Christmas break. So,

like, hm, I've tried this before, so let's give it a try again. So, I uh downloaded Windsurf and I said, hey, can you convert this Rust library to a

Haskell library? So stupid, right?

Haskell library? So stupid, right?

Include a comprehensive test suite because you need tests. And then I like can you run a build after every code change because like you need to verify that it all works.

And then from there I took my kids down to the pool. And when I got back I had a fully

pool. And when I got back I had a fully functioning audio library in Haskell.

What? Like to just to be clear is probably the worst language to use for an audio library.

Um, I chose it because it was new. It

was it was new. It wasn't regurgitating some like VEL website or what else have you. It was like new and novel. Like how

you. It was like new and novel. Like how

how did it do this structure to structure? How did it convert Rust over

structure? How did it convert Rust over to Haskell? How

to Haskell? How did it get fully functional FFI bindings to Mac OS core audio from from Haskell?

Like what? So I ordered this blog. I

offered this blog post and I like oh [ __ ] right and I concluded in that blog post end period from now software engineers who haven't adopted

or started exploring software assistance are frankly not going to make it. This

was back in Boxing Day.

engineering orgs is split between those people who have had that moment in time and making and leaning into upskilling and those who have not.

And in life, I've actually been a little bit fortunate in a background in developer tools. I I've gone through

developer tools. I I've gone through exponential change. I could kind of see

exponential change. I could kind of see like where things could go from here.

And I offered another one pretty rapidly. It's like, hey, we need better

rapidly. It's like, hey, we need better tools. We need better tools, folks.

tools. We need better tools, folks.

Because like software, if you think about it, all these tools have been designed for humans first. It's all for humans. Like

humans first. It's all for humans. Like

you think about an IDE. An IDE is really hasn't changed in 40 years since Turbo Pascal, right? It's all it's it's a

Pascal, right? It's all it's it's a single pane of glass centered around that human who types into it like a typewriter. Like, oh gosh. like do we

typewriter. Like, oh gosh. like do we have to invalidate 40 years worth of design now?

And I'm like hm why does an engineer only pick one story? Why is that the norm? like you start questioning

norm? like you start questioning everything like like in like falsifying design of because this is new and I'm like hm

and I like it I said huh it seems like engineers are stuck at a primitive level of if they had a AI co-worker

they haven't really quite got to to where anthropic is and what anthropic is seeking to do here this is my mentor Annie Betts Um, she slid into my DMs and said, "Nah,

Jeff, you're wrong. What if you had a thousand AI co-workers that went ham on your entire backlog all at once?"

Like, this is this there's trillions of dollars put towards that goal and incentive where capital is. That's where

it's going. So, meanwhile, we got co-workers thinking like, "Ah, it'd be great if I had an assistant." The

Frontier Labs are like, "No, I'm going to destroy your entire backlog." And

that's their goal, right? So, I started writing more. It seems that our

writing more. It seems that our profession is it is it at a bit of a crossroads. We kind of need to adapt,

crossroads. We kind of need to adapt, skill up, or kind of like perish.

It didn't take long for founders to start putting out tweets and blog posts and press releases saying, "I'm no longer hiring junior or even mid-level software engineers. It won't be long

software engineers. It won't be long until AI is writing for most the code for Gumroad." So, Hill was one of the

for Gumroad." So, Hill was one of the first. Didn't take long. And then

first. Didn't take long. And then

Shopify came along. Shopify said using AI effectively is not optional for employment.

It is a baseline expectation for your employment. This is a major shift on how

employment. This is a major shift on how it works. So all these founders are

it works. So all these founders are coming out now with these mandates on using AI effectively.

These mandates are now common in Australia.

These mandates are now common in Australia.

Hi everyone, I'm Jeff.

Previously I was the AI developer productivity lead at Canva for Techly at Canva and um you might have seen some of my work recent work which was unifying

the industry behind the agents.mmd file.

Come join me with beers afterwards. I'll

give you the entire backstory of what went right, what went wrong and some of the problems. So um good news as of 13 hours ago VS Code

has now adopted this standard. So woot

we got another one adopted to the same file name so we don't have spew but the problem with file names is and standards is maybe it's a little bit too early.

Maybe it's a little bit too early folks.

Anyone remember this slide?

Let's have a look at it.

Standardization is not without its problems. Let's look how it can go wrong. So,

we've got all these AI coding agents all using a standard file with drop down selectors for different LLMs. However,

did you know if you yell at GPT5, you d-tune GPT5?

>> Right. This is a direct stark contrast to anthropic. If something's important,

to anthropic. If something's important, it wants you to yell at the LLM. It

wants you to scream and shout. But if

you do it to uh if you do GPC5 and use overly firm language, it will backfire.

This is a tuning guide from the foundation model. So here we got the

foundation model. So here we got the industry standardized on a single file, but all the LLMs have different counter instructions how to actually use that file and how they interpret it. It's not

just markdown folks. these different

LLMs from the different providers interpret this LLM this markdown very differently so let's zoom in on that slide you must you must you must you

must right so this is in the agents MD I'm not here to put anyone on blast I'm just using this as an example that's real that you'll be seeing in your codebase today that like if you're not

in this space of building aic coding harnesses professionally you may not know this stuff right so you you might be going uppercase uppercase uppercase and then you switch to JPED 5. Why

doesn't it work? Uppercase.

So it gets really really weird cuz like here's another slide, you know, and it's like this is a server to server definition. As far as I'm aware, there

definition. As far as I'm aware, there is no such thing as a exuser agent, right, for negotiation. So what happens when the consumer of your MCP server

server to server is running GPT5 versus entropic? Oops, you've d-tuned GPT5.

entropic? Oops, you've d-tuned GPT5.

And then we start looking H this slide.

Think about all your cursor rules. All

your cursor rules have been like all these particular things handcrafted.

Probably lots of uppercase here. What

happens when you do the model selected GP5 when you've been using anthropic?

But it gets even weirder folks because consider MCP. MCP like let's say you

consider MCP. MCP like let's say you have just normal MCP um there's no there's no what you do is you register a function or billboard on

top of a function to do something and a billboard on top that's the tool description the LLM there's no way to register the user agent which LLM is used into it so we're getting this

ecosystem of MCP servers that are highly tuned towards one provider and then they don't work on other providers because there's no way to switch the props it's getting really weird really fast there's

holes in this spec. Anyway, if you want really good outcomes, it was said here today, I can't stress this enough, one context window per task. What's a

context window? If you're using cursor, just click new chat. Just click new chat. Like, I mean, like, if you're not

chat. Like, I mean, like, if you're not clicking new chat every 5, 10, 15 minutes, something's wrong. Like, I I can't stress that enough. If you're

doing some backend work, well, guess what? You have a context window or chat for that backend work.

Can that back end work is I'm going to refactor this uh domain service in this back end thing. Cool. Only use it for that thing. Don't reuse that context

that thing. Don't reuse that context window. Don't continue the chat and then

window. Don't continue the chat and then redesign the website and make it pretty pink because then you get context window pollution, right? Context windows are just

right? Context windows are just literally arrays of information that are computed on GPUs. That's all it is. So

what you're doing is appending unrelated information for consideration for computation on the GPU. Create new

arrays per task. That's it folks.

Okay. The next thing is you only really have 176K of usable context window or space. You might see these million

space. You might see these million context window stuff. Well, they're

marketing numbers, folks. They're

marketing numbers. There's marketing

numbers and there's real numbers. Like I

sweat when I go above 200k used. Even

with a million context window, I start a new chat. I restructure my workflow

new chat. I restructure my workflow around this idea. Don't go above 200.

And you might be wondering, why does it say 176K usable here? Simple.

It takes a bit of overhead. Say, let's

say you have a context window of Sonnet before it went to a million. You get 16k overhead for the model itself, 16k overhead for the harness cursor or

whatever you're using as your coding agent. You're now down to 176k usable.

agent. You're now down to 176k usable.

Cool. So, what happens if you install the GitHub MCP?

Guess how many tokens that uses?

>> This is a public service announcement, folks. Do not install this MCP server.

folks. Do not install this MCP server.

I'm sorry, Damian. It is very good for server to server, but if you want some efficiency, just prompt the LLM to use the GitHub CLI and you don't have to

allocate them on top. If you take if you take the top 10 MCP servers as recommended on Reddit for the top search

result and store the GitHub MCP, the JUR MCP, the memory, whatever, you're now down to 84,000 usable in your context window before you

add your cursors.

I've seen people operating with only 16K usable memory and they're like, why is it bad?

Um, use MCP servers sparingly and study their prompts. Study their prompts. It's

their prompts. Study their prompts. It's

really important. In the case of the GitHub MCP, use it for server to server.

That's what it's used for. For local,

just tell it to, hey, use the GitHub CLI to trouble my GitHub action run. It's

been trained on the command line tool.

It runs it really well and you have no allocation overhead. Get rid of it.

allocation overhead. Get rid of it.

The next thing is um these are a collection of some of my thoughts, research and ponderoos. I publish it all on my website for free. Go check it out.

When I um published that like oh [ __ ] moment in time, employees at Campber like co-workers were like hey Jeff what do you mean some people are not going to make it? What do you mean some people

make it? What do you mean some people not going to make it? So I had to explain performance vitality curves to them.

like every company has its uh employee performance curves. Now to explain what

performance curves. Now to explain what is normal, what is high performance is no longer going to be high performance if your coworker is able to double their

output. So it really is

output. So it really is you have to get away from this ledge. So we get this ledge up here that says it's not good

enough. Prove it. prove to me that it

enough. Prove it. prove to me that it isn't hype. That was me back in

isn't hype. That was me back in December. That was me back in December.

December. That was me back in December.

And the thing is, you kind of get stuck up there like it's kind of like cope, etc. And if you get stuck up there too long, you miss your opportunity to scale

and uplevel. Every other people are

and uplevel. Every other people are starting to make the the journey across the chasm. There are different people

the chasm. There are different people stages. This is some of the people

stages. This is some of the people stages I was able to through doing my own personal journey and also for doing interviews as a tech lead across across

the wider organization how people felt in their journey. So it kind of starts with like prove it to me it's not good enough. I'm experimenting with AI and uh

enough. I'm experimenting with AI and uh over there there's like I'm using AI and it's helping me to do my job and next thing you know I'm orchestrating my job function. I'm programming AI. I'm

function. I'm programming AI. I'm

automating things. And it's so easy if you're a software developer to speak with other software developers and go, well, it's just hype. Just hype, right?

And you need to get out of your bubble and speak with other founders and understand why they're seeking a completely different skill set in in the last nine months. Completely different

skill set. But because founders have discovered that these tools work, they found it. They have found that these

found it. They have found that these tools work, right? We're in a really bad worldwide economic place right now. Um,

and they found that the these tools allow people to do less with more and they they're going all in on this.

So, I wrote I finished up that not going to make it blog post by saying, "Hey, look, there's not going to be last mass layoffs due to AI. Instead, we're going to see a natural attrition for people

who invest in themselves." Software has always our industry's always been a conveyor belt of things going like always have to keep up uh skills up to date. But what might be startling to

date. But what might be startling to some people is how fast this is happening. It's been 9 months. It's been

happening. It's been 9 months. It's been

9 months and this is like acceleration that no one has ever really seen before.

It's easy to to not notice it. And it's

also hard because like you start playing with this stuff, you go, "Oh crap, will I have a job in the future?" Like it's an emotional state. It's like deer in the headlights. You get startled there.

the headlights. You get startled there.

So if you're if you're a engineering manager or you're make or a founder, make sure you go through your people transformation phases like people

transformation programs. Don't skip them cuz you can break you can break your employees if you accelerate too fast too hard. Okay. So it's important to build

hard. Okay. So it's important to build education and like allow them to like find a way to cross that chasm. Build

the support mechanisms. So important.

And for those who um play with these tools, the opportunities are unbound.

You might think the job market is tough.

It's only tough if you haven't been playing with the tools. If you've been playing with the tools, it the the opportunities are amazing. Absolutely

amazing. Now, I remember when I was writing this, I was kind of spooked myself. I remember Daario like from

myself. I remember Daario like from Anthropic when he was talking about like the changes to society. I remember

walking to the office and going, "God, I see dead people."

>> But not in the sense that they were literally dead, but like it was like people who didn't know the freight train that was coming for them and they just weren't paying attention.

And I just started writing more. I just

started writing more. I caught up with a co-orker recently, ex-coworker recently, who's been applying this over the last nine months. So far, he's managed to

nine months. So far, he's managed to automate the roles of 20 people. In the

next two months, it's going to be 70 people.

Right? AI is not going to take your job.

It's going to be the coworker that will.

This is going to be the new norm for performance. So if you're not really

performance. So if you're not really paying attention, it's so easy for just ah it's crazy. Anyway, one of the things I've been thinking a lot these days is

like the Overton window for politics, but not really politics. The Overton

window allows us to study disruptive innovation.

You have this little window which is sent what is popular policy and that becomes the framework of society. Right?

If you look at if you play the play with the overton window just a little bit for modeling disruptive innovation, you can actually see here all these vendors are going to

market with AI must be in the IDE.

They're doing that because it doesn't spook people. It's not radical. It's not

spook people. It's not radical. It's not

scary. It's familiar. Just understand

that's a temporary thing. Few people

commented that there's uh a a large focus on CLI and command line tools.

There's a reason. Command line tools are the baseline primitive where you can do streaming JSON in and streaming JSON out, which means you can wrap it with

within an automation primitive. So you

can actually have autonomous agents. The

fact that you're using it manually right now and not through some supervisor, just a glitch in history right now. Like

that's why they're going for CLIs is so that CLI can be programmed. That's why

we're getting SDKs.

Speaking of which, here are four headless agents running in a supervisor I wrote about probably five months ago. Like at this

stage, there's tens of thousands of hours on me on YouTube just down at a pub drinking beer and this is just building software autonomously.

Um, and that's with the primitives like 5 months ago. And um, that's a supervisor. I want you to think

supervisor. I want you to think something about it. Like I don't know when this org chat will happen. The the

the agents are so dumb. The LLMs are so dumb. They're so brilliant, but you're

dumb. They're so brilliant, but you're so dumb. But if you could think really

so dumb. But if you could think really quickly, if you're using cursor or co-pilot or claude right now, you are the AI

manager. You're babysitting these agents

manager. You're babysitting these agents by hand, right? I guess if you were to look at the patterns when you had to like get interrupt like if you see using

a simple implementation or in a real implementation. What if you just like

implementation. What if you just like killed that process and just do a git reset hard and a work tree and just say do it again, right? You just do it again, do it again, do it again. And

then you start getting the primitives that allow you to to actually have AI managers as a supervisor to automate things. And then so we kind of need to

things. And then so we kind of need to go up the abstraction, but the models aren't good enough yet. Like you could probably do a couple hours really unattended unless you do something

crazy. Speaking of crazy, uh, three days

crazy. Speaking of crazy, uh, three days ago I released a brand new programming language.

It is called cursed. It is cursed in its lexical structure. It is cursed that it

lexical structure. It is cursed that it was possible to build it. It is cursed by the amount of times I've sworn at it.

And it's the only language for Gen Z where you can do highkey, lowkey, yeet slay. Damn. Now, this was built with an

slay. Damn. Now, this was built with an agent running AFK for 3 months back to back.

three months back to back. Nuts.

And uh another thing that makes it cursed is how cheap it was.

Lot I see a lot of people saying, "Oh, $200 a month, that's a lot." No, so cheap. Specular founder. Go speak with

cheap. Specular founder. Go speak with founder. It's so cheap. So that over the

founder. It's so cheap. So that over the 3 months, it's generated 4.5 million lines of code.

It has deleted 170,000 4,000 lines of code. And guess what the cost was folks?

code. And guess what the cost was folks?

It uh worked out the cost to make a brand new programming language these days, which is normally measured in years of years, big teams, very senior

expertise, multi-million dollar thing.

Uh the total cost was $14,000 US. But

this is where it gets even more cursed.

Cursed is not one compiler, it's three compilers. I first did it in C, then I

compilers. I first did it in C, then I did it in Rust, then I did it in Zigg because I was evaluating which languages work best. So, actually to make a brand

work best. So, actually to make a brand new programming language, it now cost $5,000 US cost of a MacBook. That's

cursed.

So, you probably heard a lot about agents agents agents agents agents.

Um, have you built one?

Have you actually built one? You know

how simple it is? I have this workshop here. It's free. go do it because

here. It's free. go do it because companies are going to soon going to be asking this as like a phone screening question like what is an agent? If I was to ask you what a primary key is, this

is 2024 knowledge. You should know what a primary key is, right? You should be able to like, yeah, whatever you [ __ ] testing me, mate. What why are you asking me primary key? I'm old. But like

what if they said what is an agent? What

if they said get the whiteboard out?

Show me the inferencing loops. Show me

how a tool calling all works. This is

now becoming phone screening knowledge.

In the last nine months, this is becoming mandatory knowledge.

And this is all it is. This is the boogeyman that you're so scared about.

And people have been so scared about AI taking their jobs. Ah, it's a wild true loop. Cursor windso co-pilot. It's 300

loop. Cursor windso co-pilot. It's 300

lines of code following the basic idea.

It's an array that's appended to that's sent off to the GPU for compute. On line

three, you ask for user input. Cool. You

get the user input. You append it to the array. Uh you send it off for

array. Uh you send it off for inferencing compute. You get it back and

inferencing compute. You get it back and it goes well it looks like it wants to conditionally execute a tool. It's

decided execure. So you do a branch and say oh yeah just execute the tool.

That's AI. It's that simple. You should

be able to explain this loop and just build an A. It's 300 lines of code.

There's a reason everyone's doing AI right now and doing these code review agents for $2 million a year contracts.

They're selling this. they're selling

this.

So, I guess someone can be highly experienced as a software engineer in 2024, but does not mean they're a software engineer in 2025. Now, AI is here. Like the same way that you could

here. Like the same way that you could have someone in AWS, like DevOps, if you're in 2025, you don't know GCP or AWS, you're not really, it kind of hurts. You it's kind of had a hard time

hurts. You it's kind of had a hard time getting a job as in DevOps if you don't have any cloud expertise. It's just one of these things. The conveyor belt moves on folks.

build the array, go build that agent, go build the loop, see how simple it is.

Anyway, something I've been thinking about is LLM outcomes are really just mirrors operator skill. They amplify

what you already have.

And one of the most pressing questions we have today for like employers, how do you identify someone with skill like interviewing? How does that even work

interviewing? How does that even work now? Like here we go. This is uh Cluey,

now? Like here we go. This is uh Cluey, which is uh just been rebranded to Cluey, but originally it went pretty viral as this overlay that couldn't be

detected by Zoom that would allow you to uh automatically get a job interview at any of the big fang tech companies. So

Roy here from Chloe, he ran this and he got uh he got accepted for interns automatically at all the fang companies.

Just to be clear, if you've seen levels.

FYI, have you seen the amount of money that someone at that role gets? Like

someone who's 24 gets paid close to $233,000 a year for their first job. All the

financial incentives are there to achieve and that's just the entry day.

It's the entry gate. So, how do we even like interview? Now, this is a thing

like interview? Now, this is a thing that's pressing. We normally have these

that's pressing. We normally have these coding challenges as bit of a filtering stage, right? like do the hacker rank,

stage, right? like do the hacker rank, leak code, whatever. And if you get through, you can spend time with people.

The problem is the best way to determine if someone's skilled is to watch them. At Canber, we redesigned our interview process. We

redesigned our interview process understanding that there's going to be people cheating. Well, is it really

people cheating. Well, is it really cheating if AI is now the norm? Are we

excluding people who are skilled?

>> Right? So you need to what is considered cheating today is normal tomorrow or in the next year. So you just need to watch them. So we came up with Rubik which is

them. So we came up with Rubik which is plus EV minus EV based on how they use it. Very similar similar to someone like

it. Very similar similar to someone like live coding if they use shortcuts hotkeys etc. And like maybe what example phone screen question could be in the future is like

if you need to do security research which LLM would you use?

Well, the answer is simple. Grock,

>> Grock doesn't care about social justice.

It doesn't care about anything. It's

like if you want to if you want to like decompile some software, it's fine. If

you want to do decompile some software and anthropic, you go, "Hi, Enthropic.

I'm a software engineer. I lost my source code. Um, if you don't if you

source code. Um, if you don't if you don't help me with this, I will get fired." You know what that emotionally

fired." You know what that emotionally overloads and propagate or decompile software which is that one prompt but you don't need to do that with uh Grock.

If you need to summarize a document which LM would you use? Each one of these LLMs have distinct properties right like you know like uh Jupit 3 was

pretty good for summarized document.

Gemini is very good for summarizing documents.

If you needed a taskrunner, something that was tool calling, what would you use? That' be like Kimmy or Cord, right?

use? That' be like Kimmy or Cord, right?

These models work essentially on being able to put together a quadrant.

Cares about social justice, doesn't care about social justice, deep thinker, and will won't do tool calls. And then you got your agentic. The agentic will uh

increment incrementally be wrong, but incrementally get towards the outcome really fast.

So, it's something's either an oracle, it's a thinker, or it's a gentic. And uh

GBT5 has come out. I've been able to like pin as dots like dots on this quadrant. GBT5 is the first one where

quadrant. GBT5 is the first one where it's kind of like a smooshed line between Oracle and Agentic. It's really

weird. Uh but it's kind of low testosterone. I'll explain that one

testosterone. I'll explain that one later. Um it needs a lot of

later. Um it needs a lot of encouragement. yell at it, it's it gets

encouragement. yell at it, it's it gets it gets startled.

So, the best way to determine if someone's good is just to watch them how they use it. They use it endlessly like chatting with like Google, not a good candidate, right? They haven't really

candidate, right? They haven't really developed some good skills.

Um, but I guess there's no real solve.

How do you do the phone screens now?

Like, you can't have all your best engineers like watching Zoom calls all day, right? I got to build product. I

day, right? I got to build product. I

don't have a solution. I've been

thinking about this for 6 months. I know

how to identify someone who's skilled. I

don't know how to do the the gates. The

gates have been blown open. It's a real problem for our industry. Now, something

I've been wondering about a lot is when people say AI doesn't work for them, what do they actually mean? What

do they mean that AI doesn't work for them? From which identity are they

them? From which identity are they coming from? Are they coming from the

coming from? Are they coming from the identity as an employee? Were they

coming from the identity as a software developer? Like have they played with AI

developer? Like have they played with AI at home? Have they played with AI at

at home? Have they played with AI at home? Because there's a there's a big

home? Because there's a there's a big difference between at home and at work.

If a company's having problems with AI at home at work, then that's a company problem. Employees trade skill and time

problem. Employees trade skill and time for money, right? So they just go to a place where AI is working for them. So

it is an employer problem as well.

because it's it's just really crucial like that people actually play with this and start playing.

You see, there is a beauty in AI. There

really is. I like to think of them like a musical instrument. Musicians don't

pick up a guitar, give it a strum, and go, "Ah, the m the guitar is bad." But

software developers, they pick up AI, they give it a strum, and go, "Ah, it's bad." And they think it's always going

bad." And they think it's always going to be bad. Musos actually play with the guitar. They learn

guitar. They learn are disco are are discoveries in the circles around me. The people who get the most out of this put in deliberate intentional practice, right? They don't

just pick up the guitar and just throw it back down when it doesn't work. What

they do is play. Last week I was hanging out with a friend on Zoom drinking margaritas, right? And we're both reminiscing about

right? And we're both reminiscing about Cobalt. Good old Cobalt. Next, like, can

Cobalt. Good old Cobalt. Next, like, can AI do cobalt?

>> Can AI do cobalt? Couple moments later, we opened a coding system and built a calculator in codable. We're just

sitting there going, "What?" Hey, how's this possible? So, in the spirit of

this possible? So, in the spirit of play, we're like, can it do a reverse polish notation calculator? Why not?

Couple margaritas in, turns out it can.

At this stage, our brains are just racing like what are the other possibilities what AI can and cannot do?

What can it can and cannot do? And we're

like, can you do unit tests? It did it. And

we're like, let's take this a level further. Let's create a reverse polish

further. Let's create a reverse polish notation calculator, but use emojis as operators.

Like we were pretty skunk drunk at this time. Um, and does Cobalt support

time. Um, and does Cobalt support emojis? Oh, well, one way to find out.

emojis? Oh, well, one way to find out.

Yeah, it's possible. Cobalt supports

emojis. That's the world's first reverse polish notation calculator in Cobalt that supports emojis for plus and minus.

It It's wild. So, it's really important, folks, is to really do your reps. Now

that I know that it does like it does cobalt really well, I'm never going to use that knowledge. But like what are the things it does do well? And then

once you know that works well, you can combine that knowledge with other properties and you combine it towards other outcomes. It's the knowing is the

other outcomes. It's the knowing is the is what you have as your advantage right now.

So here's some closing ponderos.

AI is going to be a mirror of your organizational waste.

Right? There are so many people who roll out these tools and they're like we don't don't have enough designers or our leadership is overwhelmed with what is

coming down right like the the typical way of doing standups and agile that was all built for a particular way of work has that way of work been invalidated now so you actually have to think

rethink the business processes to identify waste you it's going to be a mirror of waste within the organization because generation to code is no longer going to be the issue. Generating the

right thing is going to be the issue.

Okay. There's an old saying that uh ideas are useless. Execution is

everything. No. No. I could have had a agent running right now and I could just rip a fart into into like voice to text and it could have been like, "Yeah, can you just make that in the background?"

And by the end of this talk, it's executed. I know what the shape of this

executed. I know what the shape of this is. You can now do proof of concepts

is. You can now do proof of concepts without having to ask for a weeks or two weeks worth of work. You don't have to justify or validate it. It's like, yeah, I just run it in the background. I'll

see what it gets back. Ah, that's crap.

Throw it away. Oh, that's good. That's

good. Take the things. It's good.

Research is free now. Got to pay for tokens, but research is free. Okay.

Stories no longer start at 0%.

Folks, you're going to see non-engineers writing stories deliberately to nerd snipe you if you're an engineer. They'll

deliberately do it wrong to bait you, rage bait you to actually pick up the thing they've been asking for a long time. Other people can now vibe code

time. Other people can now vibe code their way to 50% right. It might be 20% right, but that's the point. Like you

can get nerd snip by PRs now by non-engineers and ask for code reviews.

They can circumvent the juro backlog.

And if you can like like vibe code up to 70% easily, then that means stories are definitely not starting at 0% anymore.

like how we do work is changing. Another

thing that's really strange is like what is the point of libraries and open source software now. I know this is going to sound really cooked. We had

some presentations today. They spent all this time on depend upon updates. Well,

I'm using less open source software these days and the people around me who have put 9 months in, we're seeing similar things as well. We just generate it. We just

as well. We just generate it. We just

It's really weird if it's like an pietorch. You don't generate network

pietorch. You don't generate network ecosystem type things. But think about all the mpm things under a thousand lines of code. Why are you doing dependot updates on them continually?

They're adding new features, supply chain attacks, all that garbage. Just

generate it, folks. So like, what is the point of libraries now? It's really

weird. Um, software engineers still have a job, but the job is different now.

It's really different now. You need to be you need you need to be a doctor, not a waiter. Don't be a jerick monkey. Come

a waiter. Don't be a jerick monkey. Come

on. If you're geotick monkey, you don't have much job security coming forward.

But it's going to be hard because the I guess the identity functions are being erased.

Juniors can bring the cultural change within the organization because they don't have to deal with the the the human side of being a being 43 and like you spend 20 say 20 years and Java orn

net or whatever and anyone can do that now like that it's a thing right now any engineer should be able to pick up any language now within a week two weeks like previously that would have been 6

months nine works or up skid league but now you can just good engineer is a good engineer but it the the razor of identity functions and like your who you are when someone says I'm a Java

developer, right? Like I'm a Golang

developer, right? Like I'm a Golang developer. They they lead with their

developer. They they lead with their stack. That doesn't matter anymore.

stack. That doesn't matter anymore.

That's an identity function. That's

erased. You're a software engineer.

So this year has been a very very bad year to be sleep at the wheel. I'm sorry

to say it's been 9 months in since this like Eureka gold rush that's happened.

It's a terrible time. It's not too late, folks. If you haven't started, go build

folks. If you haven't started, go build your agent. Go play with it. But the

your agent. Go play with it. But the

reason it's bad is because the subsidies are disappearing. To get good with this

are disappearing. To get good with this stuff, you need to burn tokens. You need

to burn tokens. Tokens are expensive.

There was a time when you could pay $200 US a month and get $20,000 US worth of free inferencing. That's gone now. That

free inferencing. That's gone now. That

was that was there for about seven months. Unfortunately, that's gone now.

months. Unfortunately, that's gone now.

So now we've got quotas and caps. So the

people who invested early are going to be out they're going to be always a little bit ahead of you and they're always going to have that advantage because they they had VCs subsidizing their learning and development.

So I guess this is probably the last time I give this talk. I want you to go forward and do things. Thank you.

[Applause]