April 23, 2025 • 15 mins
Andrew Boyagi just completed a multi time zone, multi country, multi week development project with no meetings. Not one. No kick off meetings, no status meetings, zero. And no one even notices until it was pointed out after the fact.
Sure, Andrew still kicked off the project. Of course everyone kept informed. Yes, objectives and priorities were aligned and corrected over time. But without meetings. Just messages, posting, videos, and perhaps a call, using Loom, Confluence, and Slack.
As Atlassian’s Global Head of DevOps Evangelism, Andrew’s all about experience to drive better software, faster. He’s constantly thinking about removing friction, innovative use of tools and processes, and focusing on the high leverage activities that deliver. Andrew’s mission is to eliminate friction, toil, and distractions so people can focus on delivering the best work of their lives.
I had a chance to sit down with Andrew in Anaheim at Atlassian Team ‘25 and dig into the latest findings from Atlassian’s ‘System of Work.’ Andrew’s also an advocate of leveraging AI to help increase developer productivity, but perhaps not in the ways you might expect.
Please join me in welcoming Andrew to the Work 20XX podcast
Thank you Andrew
Editor's Note:
Recorded 2025-Apr-09, Anaheim Convention Center. My first solo onsite recording with some new gear. The audio is not great, but the content is well worth the effort. Missing my rock star production team.
Special shout out to Rob Castaneda and Team ServiceRocket for sharing their booth to enable this recording.
YouTube - Click Here
https://www.youtube.com/watch?v=a18HhhdVRoo&list=PLZURvMqWbYjmmJlwGj0L0jWbWdCej1Jlt
Transcript and Show Notes - Click Here
https://www.work20xx.com/episode/andrew-boyagi-better-experience-productivity-system-work-20xx-ep35
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:01):
All right, so
I’ll count us down
and we will go
in three, two, one.
Hey welcome back everybodyJeff Frick here
coming to you not from the home office
coming to you from Anaheim, California
to check in on the Atlassian Team ‘25 event
These guys are on the future of work
and have been for a long time.
So I was excited to,to roll down here
and get the latest.
So I'm excited to have my first guest
(00:22):
He’s Andrew Boyagi
He is the Global Headof DevOps Evangelism.
Andrew, great to see you.
Hey Jeff, thanks for having me.
Absolutely.
So I love people that haveevangelism in their titles.
Like, you know, on one handyou talk about
process and helping people,
but on the other,
what makes it an evangelical?
What is it is that you're outwith your message to help people
get better at basically,you like to say,
(00:44):
getting better codedeveloped quicker.
Yeah, I mean, the areas that I usuallycover with people are around
developer experience and productivity,but also helping teams
just work bettertogether as well.
You just came out with thelatest Atlassian ‘System of Work’
That comes out I think, every year.
What were some of the,
What were some of the new findingsand best practices this year?
(01:04):
So like we announced today during our keynote,
we're really focusing on helping teams,do a lot of things together, so
Tracking work against company goals about,
unlocking knowledge acrossteams and then also,
helping them to deliver work accordingto their company's priorities.
So we're focusingon a few different things there.
(01:25):
And you just
And you just highlighted that in a recent post
I saw getting ready for this,
where you talkedabout a project that you did.
I think it was a month long project
with teams all over the world,people all over the world.
While you were traveling, I thinkyou had like 3 or 4 cities.
You were traveling
and you did the whole thing without meetings.
Wow, that is amazing.
So share a little bit about,you know, kind of
(01:46):
how you decided to make that approach in the first place
to see if you could do itwithout meetings?
Yeah. I mean, I've been thinking.
I had been thinking about runninga meeting-less project for a while.
I mean, nobody likesunnecessary meetings.
But I didn't really havean opportunity,
and I hadn't put too much thought into it
about how I wouldactually do that.
And like you said, I was travelingacross multiple time zones.
I had I was working with teamsalso in multiple time zones,
(02:08):
and tobe honest,
it was hurting my braintrying to work out
how am I going to get everybody together,
to kick off a meeting?
Because usually I kick off a project,
because kicking off a project,
it’s usually something that you do in real time.
Right
It was
I was getting further and furtheraway from where we needed to be.
We had, some things,some dates that I wanted to hit
that we weren't going to hit.
Unless I kicked off the meeting pretty quickly.
(02:30):
So I wrote a few Confluence pages.
I did a kickoff pageon Confluence.
It’s a great templateyou should check that out.
And, I recorded a Loom
In the Loom, I was just giving everyone the context,
talking about whatI think we needed to do.
In the Loom, I left a lot of space for people to,
tell me what they thoughtwe needed to do as well.
Like, I was asking questions, saying,
you know, I think this is what we need to do, but,
(02:53):
you know, if you havesome expertise in your craft,
let me know what you think.
And I was good.I mean
people interacteda lot with the Loom.
Sometimes I had the wrong personor the wrong team tagged,
and they jumped in and tagged the right person in there.
And so betweenLoom, Slack and Confluence
we didn't have any meetings.
We ran the project that way.
(03:13):
Sometimes people would send a Loom back to me to say, hey,
I want to explain this thingthat we're doing
it’s too hard overa written format.
So, yeah, I mean,it went really well.
No one really even noticedthat we didn't have any meetings,
[Jeff] They didn’t even noticetil it was over
[Andrew] I was pretty annoyed about that, actually,
because I'm like thinking
we're getting towardsthe end of the project.
No meetings, no one elsescheduled any meetings either.
(03:35):
And I'm like,this is an awesome thing.
I wonder what everyone’sthinking about it.
And I did catch up witha few of the people individually.
I'm like, what do you thinkabout that meeting?
That, that project,we didn't have any meetings
and nearly all of them were like,what do you mean?
We didn't have any meeting? Like.
We did not havea meeting about it.
It was all Loom and Slack.
(03:56):
They didn't even notice
[Jeff] they didn't notice
So, I think part of that,as well as in Atlassian,
obviously we use all of our products and
We do watch a lot of Looms and
We do communicate that way so they may not have
pieced it together, but,
I know I felt like impressed with the team.
Yeah.
That they didn’tdidn’t realize,
you did it without them noticing.
I mean, there's a lot to be said for that
(04:16):
I want to ask youabout Loom and
you guys use Looma lot here at Atlassian.
And I was just in a session with Dom
was talking about using it
and then your descriptionof this project,
you talking abouta lot of meeting time is
one person kind of sharing information.
You know I can do this via Loom.
What's weird though, about Loom is
I could see people getting a little nervous
(04:36):
You know, it's one thingif you're typing
and you've got some notes,
but to actually put your face on it.
I know it conveys a ton moreof emotion and subtlety,
but do people like,
are they trying to, you know, like Instagram,
they're trying to get that perfect Loom?
How has is it been adoptedand how is it being used in terms of
it doesn't have to be perfect.
It's just a way to get a different piece of communication out.
(04:57):
You say everyone in Atlassian uses Loom a lot.
I actually started using Loombefore I joined Atlassian.
I used to use it with a nutritional studies
to work with on the other side of the world. So.
Anyway, there's lots of differentapplications for it.
But, you're right.
When I first startedreporting Looms,
when I joined Atlassian
especially when you’redoing it professionally,
I was a bit nervousthe first time,
(05:19):
and I think it took mea very long time
to record that first Loom becauseI kept rerecording it, editing it.
And so I think
I went through a process wherenow I record a Loom
I’ll make mistakes, I don't really care.It's only people I work with.
It doesn't needto be perfect, so
I think and I’ll swing to other Atlassian’s
who also go through that process
where the first one’s like
ultra high quality,like it's going to be on TV
(05:41):
and then you watch others and
slowly, slowly, you get more comfortable with
being on camerarecording,
and just sharing thingswith your peers.
I mean, if you'rein a meeting,
you can't say, oh, let's stop the meeting,
we need to get back andI need to do that again.
Right, right.
So,
But you don’t look at yourself when you're in a meeting, right?
You just are talking when you're
you're looking at that videobefore you hit send,
you know, it's that little extra pressure, right?
(06:02):
You know, look.
There's a lotof them now
I don't even watch thembefore I send them
for that reason, because it doesn't really matter
as long as I get the messageacross then it's done.
Right.
In another part of that projectyou talked about no updates
and really being very consciousabout focusing on the areas
that needed attention versusnot necessarily focusing on
and getting away from just kind ofregular, systematic updates
(06:25):
that people fall into this,this whole routine.
So how did you manage what stuffyou needed to pay attention to
and what stuff you didn't need to pay attention to?
Yeah, I mean, so I gave an update to the team every week
which is like around up of, you know,
here's where I got to
here’s where I think you're all up to as well.
Just
I used to summarizebecause we did a lot of collaboration
in Confluence and in Slack.
(06:46):
So I'd summarize
and say here’s here's where everything is at
for everyone's visibility.
When I started doing that, other people started doing it as well.
So they would startsharing an update,
every week in Slack or, you know,in various different ways.
But everyone just startedsending, an update themselves.
So I never felt likeI needed to have,
like, a statusmeeting or
(07:07):
I never really had to ask anybody
for a formal update on what was happening.
And so I thinksome of it
is leader drivenas well.
Like if a leader is saying,give me an update every week,
then you, you know,most people will start doing that.
If the leader just startssharing an update,
other people will do the same.
It's just like the way that workhappens in the modern world.
(07:27):
And I'm not familiar with Loom.
Does it auto transcribe as well,
so that you've got text searchable,
information therethat you can go back and find
and does it, does it index it
and do all those thingslike it would a normal
[Jeff] kind of typed note[Andrew] Yeah.
So it, it will do a transcript
which, we announced today, you can use right away
and get that into a Confluence page.
It's indexable.
So, it's pretty cool whatyou can do with it now.
(07:49):
[Jeff] Yeah. That's cool.
So I saw you speak,on talking about developers
and trying to help developersbe more efficient.
And you madea really interesting point
specifically how people can use AI
because there'sa lot of talk,
you know, what's the role of AI?
And you said, you know,
developers are actually pretty good coders,
and they only spend about30% of their time coding.
It's the 70% of the other crapthat they're working on.
(08:13):
And you're, you know,
kind of mission is to help peopleremove that friction.
But I thought wasreally interesting, though.
Is you found some really unique cases of AI that
that are not being used,just to develop code,
but are being usedin that other 70%
to support the processin some pretty innovative,
innovative ideas.
I wonder if you can share
some of the waysyou guys are using AI
(08:34):
to help developers,but not necessarily to write code.
Yeah, I mean, like,
there's a coding part of a developers’ job like you said
it's a minority of their time.
And so just to cover the point
that you, you also mentioned about AI
of a lot of people
when they think about AI for developers,
they're thinking about writing code
writing code faster,writing better code.
(08:55):
And, you know,
maybe AI is giving developersa small boost in that area.
But it's not a point of frictionfor developers,
so it's not going to improvedeveloper experience.
It's probably not going to vastlyimprove productivity.
And doesn't make them happier,that's for sure,
after speaking to a lot of them about that.
But there are ways you can improvedeveloper experience using AI
(09:16):
and, the processis the same.
You identify a friction point,
understand that friction point, deeply
identify a way to solve that friction point,
which may include AImay not. And,
once you remove that, you'll find that developers
are not only happier
but they are more productive as well.
A good example, for us internally,which is a friction point
(09:37):
and it’s a friction point for many people
is that PR [pull request]cycle time was taking too long.
So, developerssubmit a pull request.
People will be
they'll be waitingfor a review.
They start doing other things.
You know,if it's a week, for example,
someone does their review,gives them feedback.
Even if you'renot a developer,
you can imaginewriting a document.
You know, in a weekyou've worked on 50 other things.
(10:00):
Someone gives you feedback.
You have to go back and reread your own documents,
to work out what you wrote.
What’s this feedback obviously.
Right
You know your just, the flow is all over the place when it's like that.
So, you really want to reduce thatcycle time as much as you can.
And so we do have an AI
way of using AI to help us with that.
We have an AI agent
(10:21):
which helps with pull requests,
which essentially does,
it's like a sparring partnerfor the developer
where it'll do a first review.
Of the pull request,it'll pick up, you know,
any basic errorsthat are there.
It’ll writea description
and that way when it does
come for another humanto review that pull request,
it's already at a good standard and good quality.
(10:42):
So it makes that whole cyclea lot faster.
Right?
The part that jumped out to me
that you also said that you didn't mention here is
is helping the developers,
especially that don't have English as their first language.
[Andrew] Yeah.
To write a generalizedsummary of what they did
and that, that's just, you know,
that's where large LLMs today [Large Language Models]
just shine in that type use case.
(11:05):
I thought that waspretty innovative.
Yeah. I mean, like,there was a.
It wasn't at Atlassian,
but I was sitting in an officewith some developers,
and I had my headphones on and,
there was a developer sitting next to me,and then another one was there,
and I couldn't really hear anything.
I was listening to something,
and then eventually there was likefour people standing there.
(11:26):
And so I took off my headphones and
they were going through comments on a pull request and
how they're making some commentsabout the other person, like,
What's this guy talking about?Why is he so rude?
Like, you know, all these things,
and I don't work at that company,
but I said, can I,do you mind if I have a look?
And I had a look and
what was clear to me was
that person's first language was probably not English, right.
(11:47):
You know, I didn't thinkthere was any,
bad intent with what was written, but,
[Jeff] Right[Andrew] But
you know, it was just written in a way where,
you know, it could beperceived differently.
And so just a simple use case for AI, which is
change the tone or make it happieror make it sound friendlier
or whateverit is
can really help people who,
you know, who just,
you know we all have written an email
which has been misinterpreted before.
(12:08):
So, you know, I think that'sa nice use case as well.
Yeah. Email is just a horrible communication method.
And I remember first timeI heard about Atlassian
that you guys were not using email internally.
I don't know, ten years ago,12 years ago.
I don't know when itwhen it started
it was ten years ago that I heard about it.
And to really
it kind of speaks to reallyappreciating people's flow state
(12:31):
and really letting people be asyncand really reducing interruptions.
As I would imaginewithin that 70%
interruptions are a pretty, significant inhibitor to productivity.
I mean, I don’t use email at work
and, it's not onlyit's not only
there's afew things like
emails are just a modernized letter,
like, I don't want to send someone
(12:51):
something to someone I work with and say,
Hi, Janet.How are you today?
I hope this finds you well.blah, blah, blah. blah.
We speak to each other50 times a day.
I'm just going to say,
Hey, where did you get up to with this?
Or, you know,
Can I help you with that or whatever it is?
So it's more direct to the point.
The other thing is that
email really locks knowledgein between a few people.
Whoever hasthe email,
(13:13):
which is not helpfulwhen it comes to AI and
you would have seen a lot of thingsthat we announced around,
being able to discover informationacross an organization.
So to me, using email in that wayis an anti-pattern.
To me, I like thingswritten in documents
or somewhere like Confluence
where it’s discoverableby other people
where information’s reusable.
(13:33):
And I think that really helpsunlock other people's flow.
It's funny, once you goto shared documents
on any type of a platform,
you're like,oh my gosh,
I can't imagine version control.
And, you know,sending around revisions
and this is version 2.5.2 that Andrew took a look at that, and I have
Did you look at this one beforeAndrew took a look at it?
Yeah.
I mean it's so much,so much better.
Well even old versions in email
(13:54):
like someone attaches a document to an email
and it’s like ten revisions old.
Yeah.
And now we're just getting the threads
that go onforever and ever.
So yeah, emails is not a good thing.
So 2025,
what are you excited aboutthis year as you guys
wrap up the show and get ready for,
for, you know, you’re in such a good space
because code is only gettingmore important for enterprises.
(14:14):
And the speed ofdecision making
and rolling out is only getting faster and faster.
So what are you excited about for this year?
[Andrew] Yeah. I'm excited
about better use cases for AI,to be honest with you.
Like, we spoke about codingfor developers earlier.
But we're starting to explore more of the
the non-coding time,let's call it.
But even within that coding time,
we've got some, agentsthat we've released
(14:36):
and we're obviouslydeveloping more
but things that really helpdevelopers remove friction points,
like we do have a coding agentwhich is used for things like
removing feature flags.
So, you know, imagine an engineering manager says,
we’ve got this tech debt around.
Around, feature flags, our feature flags.
Please go and remove it.
Nobody wants to do that.Like that's toil.
(14:56):
That's friction for someone.
You know thenimagine you do three
and they're like, actuallywe have 35 more.
Can you go and do those as well?
So, you know,
I'm really lookingforward to
I feel like last year was
a lot of people learning about AI and
and finding new use cases.
I think this year,
we're going to see some of thosereally come to life.
[Jeff] Awesome.
Well, Andrew, thanks fortaking a minute.
I know you're busy.Congrats on the show.
(15:18):
And, really great event.
Thanks, Jeff. Appreciate it.
[Jeff] All right. [Andrew] Thank you.
He's AndrewI'm Jeff,
you're watching Work 20XX special edition
from the Atlassian Team ‘25 showdown in Anaheim, California.
Thanks for watching.Catch you next time.
All right.
Awesome.
We’re good Thank you.
Thank you.
Hey, Jeff Frick Here
big shout out to the podcast audience.
(15:38):
Thanks for listening in.
You can get show notes and transcripts at Work20XX.com
And that also has links to the videos as well.
Appreciate you listeningin on the podcast
Do reach out
say hello, like subscribeand smash that notification bell.
Thanks for listening.Take care. Bye bye.
All right, so
I’ll count us down
and we will go
in three, two, one.
Hey welcome back everybodyJeff Frick here
coming to you not from the home office
coming to you from Anaheim, California
to check in on the Atlassian Team ‘25 event
These guys are on the future of work
and have been for a long time.
So I was excited to,to roll down here
and get the latest.
So I'm excited to have my first guest
(00:22):
He’s Andrew Boyagi
He is the Global Headof DevOps Evangelism.
Andrew, great to see you.
Hey Jeff, thanks for having me.
Absolutely.
So I love people that haveevangelism in their titles.
Like, you know, on one handyou talk about
process and helping people,
but on the other,
what makes it an evangelical?
What is it is that you're outwith your message to help people
get better at basically,you like to say,
(00:44):
getting better codedeveloped quicker.
Yeah, I mean, the areas that I usuallycover with people are around
developer experience and productivity,but also helping teams
just work bettertogether as well.
You just came out with thelatest Atlassian ‘System of Work’
That comes out I think, every year.
What were some of the,
What were some of the new findingsand best practices this year?
(01:04):
So like we announced today during our keynote,
we're really focusing on helping teams,do a lot of things together, so
Tracking work against company goals about,
unlocking knowledge acrossteams and then also,
helping them to deliver work accordingto their company's priorities.
So we're focusingon a few different things there.
(01:25):
And you just
And you just highlighted that in a recent post
I saw getting ready for this,
where you talkedabout a project that you did.
I think it was a month long project
with teams all over the world,people all over the world.
While you were traveling, I thinkyou had like 3 or 4 cities.
You were traveling
and you did the whole thing without meetings.
Wow, that is amazing.
So share a little bit about,you know, kind of
(01:46):
how you decided to make that approach in the first place
to see if you could do itwithout meetings?
Yeah. I mean, I've been thinking.
I had been thinking about runninga meeting-less project for a while.
I mean, nobody likesunnecessary meetings.
But I didn't really havean opportunity,
and I hadn't put too much thought into it
about how I wouldactually do that.
And like you said, I was travelingacross multiple time zones.
I had I was working with teamsalso in multiple time zones,
(02:08):
and tobe honest,
it was hurting my braintrying to work out
how am I going to get everybody together,
to kick off a meeting?
Because usually I kick off a project,
because kicking off a project,
it’s usually something that you do in real time.
Right
It was
I was getting further and furtheraway from where we needed to be.
We had, some things,some dates that I wanted to hit
that we weren't going to hit.
Unless I kicked off the meeting pretty quickly.
(02:30):
So I wrote a few Confluence pages.
I did a kickoff pageon Confluence.
It’s a great templateyou should check that out.
And, I recorded a Loom
In the Loom, I was just giving everyone the context,
talking about whatI think we needed to do.
In the Loom, I left a lot of space for people to,
tell me what they thoughtwe needed to do as well.
Like, I was asking questions, saying,
you know, I think this is what we need to do, but,
(02:53):
you know, if you havesome expertise in your craft,
let me know what you think.
And I was good.I mean
people interacteda lot with the Loom.
Sometimes I had the wrong personor the wrong team tagged,
and they jumped in and tagged the right person in there.
And so betweenLoom, Slack and Confluence
we didn't have any meetings.
We ran the project that way.
(03:13):
Sometimes people would send a Loom back to me to say, hey,
I want to explain this thingthat we're doing
it’s too hard overa written format.
So, yeah, I mean,it went really well.
No one really even noticedthat we didn't have any meetings,
[Jeff] They didn’t even noticetil it was over
[Andrew] I was pretty annoyed about that, actually,
because I'm like thinking
we're getting towardsthe end of the project.
No meetings, no one elsescheduled any meetings either.
(03:35):
And I'm like,this is an awesome thing.
I wonder what everyone’sthinking about it.
And I did catch up witha few of the people individually.
I'm like, what do you thinkabout that meeting?
That, that project,we didn't have any meetings
and nearly all of them were like,what do you mean?
We didn't have any meeting? Like.
We did not havea meeting about it.
It was all Loom and Slack.
(03:56):
They didn't even notice
[Jeff] they didn't notice
So, I think part of that,as well as in Atlassian,
obviously we use all of our products and
We do watch a lot of Looms and
We do communicate that way so they may not have
pieced it together, but,
I know I felt like impressed with the team.
Yeah.
That they didn’tdidn’t realize,
you did it without them noticing.
I mean, there's a lot to be said for that
(04:16):
I want to ask youabout Loom and
you guys use Looma lot here at Atlassian.
And I was just in a session with Dom
was talking about using it
and then your descriptionof this project,
you talking abouta lot of meeting time is
one person kind of sharing information.
You know I can do this via Loom.
What's weird though, about Loom is
I could see people getting a little nervous
(04:36):
You know, it's one thingif you're typing
and you've got some notes,
but to actually put your face on it.
I know it conveys a ton moreof emotion and subtlety,
but do people like,
are they trying to, you know, like Instagram,
they're trying to get that perfect Loom?
How has is it been adoptedand how is it being used in terms of
it doesn't have to be perfect.
It's just a way to get a different piece of communication out.
(04:57):
You say everyone in Atlassian uses Loom a lot.
I actually started using Loombefore I joined Atlassian.
I used to use it with a nutritional studies
to work with on the other side of the world. So.
Anyway, there's lots of differentapplications for it.
But, you're right.
When I first startedreporting Looms,
when I joined Atlassian
especially when you’redoing it professionally,
I was a bit nervousthe first time,
(05:19):
and I think it took mea very long time
to record that first Loom becauseI kept rerecording it, editing it.
And so I think
I went through a process wherenow I record a Loom
I’ll make mistakes, I don't really care.It's only people I work with.
It doesn't needto be perfect, so
I think and I’ll swing to other Atlassian’s
who also go through that process
where the first one’s like
ultra high quality,like it's going to be on TV
(05:41):
and then you watch others and
slowly, slowly, you get more comfortable with
being on camerarecording,
and just sharing thingswith your peers.
I mean, if you'rein a meeting,
you can't say, oh, let's stop the meeting,
we need to get back andI need to do that again.
Right, right.
So,
But you don’t look at yourself when you're in a meeting, right?
You just are talking when you're
you're looking at that videobefore you hit send,
you know, it's that little extra pressure, right?
(06:02):
You know, look.
There's a lotof them now
I don't even watch thembefore I send them
for that reason, because it doesn't really matter
as long as I get the messageacross then it's done.
Right.
In another part of that projectyou talked about no updates
and really being very consciousabout focusing on the areas
that needed attention versusnot necessarily focusing on
and getting away from just kind ofregular, systematic updates
(06:25):
that people fall into this,this whole routine.
So how did you manage what stuffyou needed to pay attention to
and what stuff you didn't need to pay attention to?
Yeah, I mean, so I gave an update to the team every week
which is like around up of, you know,
here's where I got to
here’s where I think you're all up to as well.
Just
I used to summarizebecause we did a lot of collaboration
in Confluence and in Slack.
(06:46):
So I'd summarize
and say here’s here's where everything is at
for everyone's visibility.
When I started doing that, other people started doing it as well.
So they would startsharing an update,
every week in Slack or, you know,in various different ways.
But everyone just startedsending, an update themselves.
So I never felt likeI needed to have,
like, a statusmeeting or
(07:07):
I never really had to ask anybody
for a formal update on what was happening.
And so I thinksome of it
is leader drivenas well.
Like if a leader is saying,give me an update every week,
then you, you know,most people will start doing that.
If the leader just startssharing an update,
other people will do the same.
It's just like the way that workhappens in the modern world.
(07:27):
And I'm not familiar with Loom.
Does it auto transcribe as well,
so that you've got text searchable,
information therethat you can go back and find
and does it, does it index it
and do all those thingslike it would a normal
[Jeff] kind of typed note[Andrew] Yeah.
So it, it will do a transcript
which, we announced today, you can use right away
and get that into a Confluence page.
It's indexable.
So, it's pretty cool whatyou can do with it now.
(07:49):
[Jeff] Yeah. That's cool.
So I saw you speak,on talking about developers
and trying to help developersbe more efficient.
And you madea really interesting point
specifically how people can use AI
because there'sa lot of talk,
you know, what's the role of AI?
And you said, you know,
developers are actually pretty good coders,
and they only spend about30% of their time coding.
It's the 70% of the other crapthat they're working on.
(08:13):
And you're, you know,
kind of mission is to help peopleremove that friction.
But I thought wasreally interesting, though.
Is you found some really unique cases of AI that
that are not being used,just to develop code,
but are being usedin that other 70%
to support the processin some pretty innovative,
innovative ideas.
I wonder if you can share
some of the waysyou guys are using AI
(08:34):
to help developers,but not necessarily to write code.
Yeah, I mean, like,
there's a coding part of a developers’ job like you said
it's a minority of their time.
And so just to cover the point
that you, you also mentioned about AI
of a lot of people
when they think about AI for developers,
they're thinking about writing code
writing code faster,writing better code.
(08:55):
And, you know,
maybe AI is giving developersa small boost in that area.
But it's not a point of frictionfor developers,
so it's not going to improvedeveloper experience.
It's probably not going to vastlyimprove productivity.
And doesn't make them happier,that's for sure,
after speaking to a lot of them about that.
But there are ways you can improvedeveloper experience using AI
(09:16):
and, the processis the same.
You identify a friction point,
understand that friction point, deeply
identify a way to solve that friction point,
which may include AImay not. And,
once you remove that, you'll find that developers
are not only happier
but they are more productive as well.
A good example, for us internally,which is a friction point
(09:37):
and it’s a friction point for many people
is that PR [pull request]cycle time was taking too long.
So, developerssubmit a pull request.
People will be
they'll be waitingfor a review.
They start doing other things.
You know,if it's a week, for example,
someone does their review,gives them feedback.
Even if you'renot a developer,
you can imaginewriting a document.
You know, in a weekyou've worked on 50 other things.
(10:00):
Someone gives you feedback.
You have to go back and reread your own documents,
to work out what you wrote.
What’s this feedback obviously.
Right
You know your just, the flow is all over the place when it's like that.
So, you really want to reduce thatcycle time as much as you can.
And so we do have an AI
way of using AI to help us with that.
We have an AI agent
(10:21):
which helps with pull requests,
which essentially does,
it's like a sparring partnerfor the developer
where it'll do a first review.
Of the pull request,it'll pick up, you know,
any basic errorsthat are there.
It’ll writea description
and that way when it does
come for another humanto review that pull request,
it's already at a good standard and good quality.
(10:42):
So it makes that whole cyclea lot faster.
Right?
The part that jumped out to me
that you also said that you didn't mention here is
is helping the developers,
especially that don't have English as their first language.
[Andrew] Yeah.
To write a generalizedsummary of what they did
and that, that's just, you know,
that's where large LLMs today [Large Language Models]
just shine in that type use case.
(11:05):
I thought that waspretty innovative.
Yeah. I mean, like,there was a.
It wasn't at Atlassian,
but I was sitting in an officewith some developers,
and I had my headphones on and,
there was a developer sitting next to me,and then another one was there,
and I couldn't really hear anything.
I was listening to something,
and then eventually there was likefour people standing there.
(11:26):
And so I took off my headphones and
they were going through comments on a pull request and
how they're making some commentsabout the other person, like,
What's this guy talking about?Why is he so rude?
Like, you know, all these things,
and I don't work at that company,
but I said, can I,do you mind if I have a look?
And I had a look and
what was clear to me was
that person's first language was probably not English, right.
(11:47):
You know, I didn't thinkthere was any,
bad intent with what was written, but,
[Jeff] Right[Andrew] But
you know, it was just written in a way where,
you know, it could beperceived differently.
And so just a simple use case for AI, which is
change the tone or make it happieror make it sound friendlier
or whateverit is
can really help people who,
you know, who just,
you know we all have written an email
which has been misinterpreted before.
(12:08):
So, you know, I think that'sa nice use case as well.
Yeah. Email is just a horrible communication method.
And I remember first timeI heard about Atlassian
that you guys were not using email internally.
I don't know, ten years ago,12 years ago.
I don't know when itwhen it started
it was ten years ago that I heard about it.
And to really
it kind of speaks to reallyappreciating people's flow state
(12:31):
and really letting people be asyncand really reducing interruptions.
As I would imaginewithin that 70%
interruptions are a pretty, significant inhibitor to productivity.
I mean, I don’t use email at work
and, it's not onlyit's not only
there's afew things like
emails are just a modernized letter,
like, I don't want to send someone
(12:51):
something to someone I work with and say,
Hi, Janet.How are you today?
I hope this finds you well.blah, blah, blah. blah.
We speak to each other50 times a day.
I'm just going to say,
Hey, where did you get up to with this?
Or, you know,
Can I help you with that or whatever it is?
So it's more direct to the point.
The other thing is that
email really locks knowledgein between a few people.
Whoever hasthe email,
(13:13):
which is not helpfulwhen it comes to AI and
you would have seen a lot of thingsthat we announced around,
being able to discover informationacross an organization.
So to me, using email in that wayis an anti-pattern.
To me, I like thingswritten in documents
or somewhere like Confluence
where it’s discoverableby other people
where information’s reusable.
(13:33):
And I think that really helpsunlock other people's flow.
It's funny, once you goto shared documents
on any type of a platform,
you're like,oh my gosh,
I can't imagine version control.
And, you know,sending around revisions
and this is version 2.5.2 that Andrew took a look at that, and I have
Did you look at this one beforeAndrew took a look at it?
Yeah.
I mean it's so much,so much better.
Well even old versions in email
(13:54):
like someone attaches a document to an email
and it’s like ten revisions old.
Yeah.
And now we're just getting the threads
that go onforever and ever.
So yeah, emails is not a good thing.
So 2025,
what are you excited aboutthis year as you guys
wrap up the show and get ready for,
for, you know, you’re in such a good space
because code is only gettingmore important for enterprises.
(14:14):
And the speed ofdecision making
and rolling out is only getting faster and faster.
So what are you excited about for this year?
[Andrew] Yeah. I'm excited
about better use cases for AI,to be honest with you.
Like, we spoke about codingfor developers earlier.
But we're starting to explore more of the
the non-coding time,let's call it.
But even within that coding time,
we've got some, agentsthat we've released
(14:36):
and we're obviouslydeveloping more
but things that really helpdevelopers remove friction points,
like we do have a coding agentwhich is used for things like
removing feature flags.
So, you know, imagine an engineering manager says,
we’ve got this tech debt around.
Around, feature flags, our feature flags.
Please go and remove it.
Nobody wants to do that.Like that's toil.
(14:56):
That's friction for someone.
You know thenimagine you do three
and they're like, actuallywe have 35 more.
Can you go and do those as well?
So, you know,
I'm really lookingforward to
I feel like last year was
a lot of people learning about AI and
and finding new use cases.
I think this year,
we're going to see some of thosereally come to life.
[Jeff] Awesome.
Well, Andrew, thanks fortaking a minute.
I know you're busy.Congrats on the show.
(15:18):
And, really great event.
Thanks, Jeff. Appreciate it.
[Jeff] All right. [Andrew] Thank you.
He's AndrewI'm Jeff,
you're watching Work 20XX special edition
from the Atlassian Team ‘25 showdown in Anaheim, California.
Thanks for watching.Catch you next time.
All right.
Awesome.
We’re good Thank you.
Thank you.
Hey, Jeff Frick Here
big shout out to the podcast audience.
(15:38):
Thanks for listening in.
You can get show notes and transcripts at Work20XX.com
And that also has links to the videos as well.
Appreciate you listeningin on the podcast
Do reach out
say hello, like subscribeand smash that notification bell.
Thanks for listening.Take care. Bye bye.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com