Episode 08-1 Transcript

Episode 8-1 Transcript

Welcome back to the main thread.
Our guest today is Lee mckeeman who is currently a staff engineer at Google,
but he's also spent time at Amazon and Meta.
I've been connected with Lee on linkedin for a while and he is prolific.
He shares openly about past mistakes,
learnings and hilarious anecdotes from his life in tech.
He also shares observations about what makes for a successful inclusive work environment.
And it's this last point that made us want to invite him on the show.
Lee is passionate about how we as technical leaders can influence our work environments to be more empathetic and inclusive so that people of varying races,
genders,
creeds,
ages and neuro diversities all feel empowered to contribute their best ideas to the work that we do.
We enjoy talking with Lee so much that we went way over our usual time allotment.
So what we're going to do is release this episode in two parts,
one today and one next week.
So that's enough out of me.
Let's get into part one of our conversation with Lee mckeeman on empathy and inclusion.
Welcome back to the main thread,
everybody.
I'm Brian Ogilvy and I'm Alex Kaiser and we are joined today by Lee mckeeman who is a staff engineer in gmail notifications at Google.
Uh Lee,
thanks so much for being on the show.
Welcome.
Absolutely happy to be here.
Um Yeah,
I wanted to give some background about myself and then we can sort of get into a conversation.
So,
um I have been at Google for about 18 months.
Um I've worked,
you know,
all in the workspace area which are the sort of productivity products.
Um the first year or so was working in on the chat side,
um working on notifications there as well as the sort of spaces which are the the chat rooms essentially um in the Google Chat product.
And um while I was working,
they sort of,
you know,
got interested in the notification side.
Um The notifications infrastructure was shared between chat and Gmail and we ultimately found that we needed a little bit different focus.
So moved some folks over specifically into the Gmail organ,
including myself.
So now we're sort of focusing up on that both on infrastructure level and um you know,
ideally some product improvements as well,
we wanna make notifications as sort of useful and delightful as we can.
And,
you know,
a lot of times people are thinking like,
well,
email is like old,
where's my like quick app?
How do I,
you know why,
why am I interacting with email?
Isn't that just for my spam and signing up for websites and we want it to be more useful and if we,
you know,
know that you get packages and we know that you,
you know,
have things,
you know,
trips coming up.
Why can't we be smarter?
So it's something we,
we're aspiring to um also trying to like,
you know,
keep our infrastructure in good shape,
use the,
you know,
built in Google smarts that we have both for,
you know,
infrastructure and high volume as well as whatever A I things make sense.
And,
you know,
right now for notifications,
we're not talking about generative A I because normally you're not being notified about something that happened already,
something that actually happened in reality.
Right.
Sure.
But it,
it may be possible to,
to leverage machine learning to say like what is relevant,
like instead of someone,
you know,
picking and choosing exactly what tags they want to be notified for.
Can we say things I haven't responded to and have,
you know,
some rules,
whether that's an actual ML model or just voyeuristic that we could be able to,
you know,
do a better job with that.
So,
um before I joined Google,
I spent a year at Meta,
um this was all sort of during pandemic lockdown um and didn't,
didn't have a,
a great time there.
It's not the fault of anyone in particular.
But um it was a big adjustment.
Um prior to that,
I was at Amazon for about nine years and coming into a new company during the pandemic,
full remote,
never made my way out there,
which I sort of regret,
um,
out there being like Seattle where the,
the rest of my team was,
oh,
you were fully remote the whole time.
Yeah.
And,
and,
you know,
the full remote onboarding process,
they were still,
you know,
trying to get that figured out team matching when you're not really sitting with the team,
like all,
all of the processes,
you know,
they were aspiring for those to be great and as good for remote folk.
And,
and for me,
I,
I sort of struggled with that.
So,
um in any event,
good experience um shook up what I had been used to at Amazon.
But um it's been a little bit of a smoother process at Google,
but,
you know,
I've also had those learnings and was able to do that differently.
You know,
both of these roles have been sort of e six staff engineer roles at Meta and Google.
Prior to that at Amazon,
I worked as a people manager um at Amazon.
There's not the sort of,
you know,
TLM like you can be an individual contributor but have a few reports and sort of do that hybrid thing.
You're either an engineer or you're a manager at Amazon and that's,
you know,
you have to choose.
And so I had chosen to move into a management role and had been managing a team that ranged in size while I was managing them from like,
3 to 7 or eight.
I was in a meeting today with someone,
uh,
who was talking about having gone through that TLM track and just feeling as though they were doing both things poorly,
you know,
I think that's a real tough job to do.
Well,
although there's a lot of similarities,
I think,
er,
when you get to that staff level between where you still are an individual contributor,
but you're expected to do quite a lot of managing of,
at least people's emotions and expectations.
Yeah.
Absolutely.
And I think that that was something that I,
you know,
learned being a people manager is like,
what I thought was that this,
you know,
I'm gonna be able to really focus on people's career development and I'm gonna get them promoted and I'm gonna teach them and it's,
you know,
it's gonna be great.
I'm gonna be able to just focus all my time on that and it,
you know,
turned out,
uh,
a little bit differently than that.
You don't have all of your time available as an engineering manager to focus on people's career dev,
like it is an aspect for sure.
Um,
and a responsibility but you have to plan team strategy,
you have to sort of manage out and up and down,
like there's just a lot expected um in that role and some of it is,
you know,
running interference for your engineers so they can get things done and keeping them from some distractions and that can be,
you know,
depending on how many distractions a full time job on its own.
So,
um you know,
I,
I moved to that role from an engineering role and I had sort of been kind of a lead engineer.
I never made it to senior engineer at Amazon,
but was kind of always working for the senior manager and responsible across teams for other engineers.
And I was like,
oh,
well,
I guess the natural progression is to move into people management.
Um And after a few years of that,
I kind of realized maybe working as an engineer,
but being in a sort of more senior role where I'm still in the room,
essentially when decisions are being made or to advocate on behalf of engineers in the org.
But I'm not having to manage them directly and,
you know,
essentially manage projects or manage expectations or manage a timeline and all of those things as,
as directly as I have to sort of as an engineering manager.
Very cool.
And that's actually a great transition into our topic for the day,
which is creating an environment of diversity and inclusion as well as an environment in which people can grow their careers.
And since you've worn both hats,
you're kind of the perfect person to talk about that.
Um Why don't you tell us a little bit about how you created that sort of environment,
an environment of diversity and inclusion,
as well as AAA growth mindset environment for all of your,
for,
for the teams that you've worked with.
Sure.
Yeah.
So,
I mean,
this is something that I'm really passionate about.
I've,
you know,
found myself feeling sort of outside in a lot of situations in life and have tried to overcome that myself.
But then I'm particularly sensitive to that when I see that happening to others.
And especially when it's,
you know,
most of the time it,
it,
it is incidental or unintentional or,
you know,
an extension of something systemic and not people being hateful or whatever.
But,
you know,
when I see that,
why doesn't this team have any women on it?
Why has it never had any women on it?
Or like there's just sort of a monoculture here,
like can the best decisions be made and users be advocated for in the best way if everyone was born in the US and went to a US University and got their four year CS degree and did an internship at a big company and started working there and like,
how,
how varied of experiences can you have?
And certainly people do,
but it is different if everyone is right handed and everyone is a tall white male or a South Asian male.
Like,
like how many different opinions and experiences can you bring to the table?
And for me,
the more the better and that's sort of proven out in studies and we see all the time that it's like,
well,
why did that camera never take pictures?
Right.
For certain users,
like,
and it's like,
oh,
because they hold it with their left hand because they're left handed and all of your engineers and Q A people and whatever were right handed.
So they always tilted it a certain way.
So you assumed up was always the right hand side of the screen and it's like,
all right.
Yeah.
So for me,
you know,
some of the,
the specifics,
you know,
I personally,
and,
and we can talk more about this later on.
I'm autistic and uh you know,
really try to champion some neurodiversity,
you know,
when I was working as a bar raiser at Amazon,
which is a person who sort of facilitates hiring decisions in the interview process.
Um you know,
would have to help dispel some of these sort of built in biases that people may not know about and everyone took the anti bias training and,
you know,
the unconscious bias,
like everyone was aware but still might not know that.
Oh,
hey,
that actually is an example of this.
So if someone says,
oh,
well,
they were really,
you know,
they,
they didn't,
didn't make a lot of eye contact or they,
you know,
kind of talked more slowly or a little bit low,
like,
I don't know how engaged they were,
be like,
hold on like I,
I don't think that that's,
you know,
a,
a reason we should take a negative opinion of the candidate.
It is possible that this person is being more thoughtful or it's possible that this is the result of a neuro difference,
is this going to affect their ability to do the job.
And even if it is,
can that be overcome with an accommodation?
Like,
you know,
we shouldn't use that as a,
a judgment that,
oh this person's a bad fit because,
you know,
they,
they didn't look me in the eye and you know,
trying to dispel some of those traditional,
like this is professional,
like professional is often coding for is what we're used to and is a product of a white patriarchy.
And so trying to not do that is super important to me.
Absolutely.
Yeah.
You know,
I Lee,
we've been connected on linkedin for a while.
I've been following you for longer than that.
Um And I'm always very intrigued by the content that you post and you post a lot.
Um But I'm always intrigued by the way that you write,
that seems to shift perspective in a way that I don't find a lot of other content creators on linkedin do.
Uh And part of it is because you are so open about yourself,
your past,
you're open about the ways in which you feel different.
You're open about past failures.
And I think that it creates an environment where you seem to be someone who is trustworthy,
who is honest and who is open.
And I think that talking openly is one of the most important first steps to creating an inclusive environment.
But I think when you get past that,
there's often a lot of well meaning people who do not experience a particular uh difference,
whether it's neurodiversity or being of a non minority member of say a gender or sexual identity or um or race,
there's a lot of,
well,
meaning people who just fail to be the allies that they need to be.
And I'm wondering if you can talk about what are the common ways that,
well,
meaning people fail to be the allies that,
that are needed.
Yeah,
sure.
So,
uh like you said,
well,
meaning is,
is the key word and assuming positive intent is I,
you know,
a really important thing and that doesn't mean we don't call people out and,
you know,
it doesn't have to be shameful.
It doesn't have to be mean spirited when someone makes a misstep.
Now,
I certainly understand if someone is,
you know,
part of a minority group or otherwise sort of marginalized person that they,
they may have that negative reaction.
And I totally understand that.
But,
you know,
me as you know,
otherwise,
uh you know,
the people that are listening to this on audio don't see me,
but I'm a tall white male that's born in America and a native English speaker.
And cis gendered and heterosexual and like,
you know,
essentially jackpot in terms of privilege.
Um I feel like I have to use that like,
yes,
the,
the autism piece is the sort of difference,
but in general terms,
have checked all the boxes for having the sort of easiest road in this country.
So with that,
if,
if someone,
you know,
makes a mistake or,
you know,
doesn't even know they made a mistake,
but,
you know,
speaks in a way that might be insensitive or,
you know,
not inclusive.
I try to approach that depending on what it is.
You know,
if it,
if it's a general statement to a group be like,
oh,
you know,
actually ii I don't think that that makes sense if people are like,
oh,
we need everyone to have their cameras on all the time.
It's like,
oh,
well,
not everyone is,
is safe or comfortable doing that.
So I think it,
it's cool for us to be able to see everyone.
Um and maybe we all say hello with video on at the beginning of the call.
But if people,
you know,
need to be fidgeting or need to be eating or can't keep their face in a presentable position for all of the call,
um we need to,
you know,
be,
be cool with everyone turning your camera off.
So if that's the need you have like go for it and just sort of in some cases supersede what a you know,
someone running a meeting might say,
because they're not trying to hurt people that might be neuro diverse or hurt people that may,
you know,
just have some anxiety about that.
And,
you know,
if it's presented as well,
here's some counter examples,
here's some reasons that are perfectly fine that aren't someone's playing their switch,
that they might have their camera off and then people can sort of become aware of it and then ideally think about that in the future.
So,
you know,
from this position,
it,
it is mostly trying to say,
hey,
I assume that there was nothing meant by this.
But you know,
brown bag can be offensive to people like a a brown bag test of someone's skin to see if they count as white or black.
Not a lot of people are aware of that except maybe people who are mixed race or who are black that are like,
hm.
And it might just be that like,
uh and they may not feel like,
you know,
this one,
I,
I don't,
I don't feel like fighting this one like and so trying to be pretty outspoken a again acting from a position of privilege like,
I'm a staff engineer,
I'm a very senior member of the team.
I I don't have a lot to risk in calling these things out.
And so doing so I feel like is kind of a responsibility.
Yeah,
that's really important.
I think a lot of times the,
the person who's having that feeling is oftentimes the person who feels the least empowered to say something about it.
And that's why it's so important that someone else who has that more sense of privilege or sense of power in the situation is the one to,
to step out and say something so that they can give power to someone else who doesn't feel like they have it at the moment,
right?
And why it can be difficult with somebody who has all the power to go around and say like,
oh everything should be just be better.
But if you don't have,
but if you aren't empowering the people around you to actually speak up and you're not receptive to that,
then that can in many ways not even make any difference at all,
maybe even do more harm than good.
Sure.
And you can't demand psychological safety,
right?
Like exactly isn't possible.
It's an environment that you have to create,
which is why I wanted to talk about the other main point of this,
of this uh episode which is empathy and I feel like empathy is like maybe the core to all of this to like communication um communication and empathy I think are like the core to this because you,
if you don't create an environment where communication is open and it's empathetic,
then people will not feel safe um to either voice concerns or to push back on concerns that you know,
push back on concerns.
Right?
You can't have either way,
either one of those,
you can't have honest exchange of ideas without that open feeling of safe communication,
which actually coincides with a lot of the other things that we've learned about what makes a good staff engineer that you need to be the sort of person that fo that environment in all areas.
Yeah.
And you know,
I think that from that perspective of empathy,
I think that,
you know,
in terms of creating an inclusive and a psychologically safe environment,
you can teach people very quickly that they need to stop talking.
Like it can be a one time thing of like dismissing someone's input or talking over someone when they try to make their point or ending a meeting.
Yeah,
it is the top of the hour.
But someone was like,
really try like they'd been trying the whole meeting to get something spoken and that's enough for some people.
And again,
especially a lot of marginalized people,
it's like,
ok,
message received and potentially that wasn't the message that was,
you know,
intended.
But that's where we have to like,
think about that and thinking about how that could feel bad,
you know,
that the two sides of the communication,
like what I said is different than what I intended and what was heard was potentially different than what I said and different from what I intended and then how that felt to that person and what,
you know,
how they interpreted it is,
you know,
1/4 possibility.
And,
and so yeah,
based on that,
like checking in can be really important and really uncomfortable.
Like after a particularly heated meeting,
maybe we were all really passionate about this discussion about this,
you know,
potential change in our architecture or maybe someone felt like they were really being railroaded and not listened to or that,
you know,
when people were raising their voice,
that was putting them into a really bad place.
And I'm not saying there's never a time,
but generally we don't need to raise our voices.
Sometimes people would to be heard because there's a loud din in the room or whatever.
But um checking in and be like,
hey,
like,
is that,
was that all right?
Like,
did that go?
OK.
And if not,
like I,
is there a way that we could do that differently?
Like,
do we need to have a different way?
We acknowledge people to speak both in,
you know,
a hybrid environment,
both in the room and on the VC?
Like,
do we need to moderate that differently because it,
it,
you know,
I,
I don't know,
but it felt like maybe you were kind of not getting to,
to speak in the way that you wanted to.
And if they're like,
oh no,
it was cool.
And I followed up with person afterwards and,
you know,
it,
it's not a big deal and I don't think we need to change anything.
I,
I can,
you know,
assert myself a little bit differently the next time.
Ok.
Cool.
No harm,
no foul.
And if it's like,
yeah,
this is like the fourth time that I've had a meeting like this where,
like,
because I'm not the loudest person,
I don't feel like I get to as assert my opinions and it,
it really kind of makes me feel like they don't matter.
So I,
I,
I'm kind of burning out on that.
I don't really want to do that more.
It's like,
oh,
all right.
Well,
let's try to fix that right now.
It's very delicate,
that level of trust that,
to foster that environment.
It's,
it's way more delicate than people I think realize.
Um,
and people will often take no news is good news.
If they're not actively complaining,
then maybe they're not,
that other person is not,
doesn't have a problem,
but it could just be that you've broken their trust and they don't feel at all comfortable.
And sometimes the first communication you get is their two week notice,
like,
or their no week notice because they're kind of totally fed up.
Exactly.
Yeah.
Yeah.
Exactly.
Yeah.
So,
Lee,
I wondered if we could,
um,
I'm curious if we could transition this to how you take this mindset for diversity and inclusion that you bring and fos you're in your workplace.
And how do you put that into the product that you're working on,
how do you advocate for diversity,
inclusion of people who actually already been in the room with you people that you think will be using what you build down the road?
Sure.
So I mean,
this,
this can be really challenging because again,
II I like to have diverse perspectives,
but as you noted,
not everyone's gonna be in the room even if you have some gender diversity and you have some,
you know,
country of origin diversity,
like you're not gonna get all of it.
And so then it's a question of,
are you asking users,
can you ask users?
Is that a possibility if you can't?
I feel like what I learn new things,
new ways that we as an industry have messed up all the time.
It's like,
oh that seemed cute and was offensive or like,
oh that like,
oh yeah,
it's cool when you click a button and the color changes and it's like uh well,
20% of people can't see that color change.
So maybe we need the text to change or an icon to change as well or something else,
then button turns from green to red to indicate that something has happened or that you need to click one more time to confirm it.
It's like,
why would,
why would someone just click the button twice as,
as far as they're concerned?
Nothing changed about it.
Just out of frustration.
They might click it multiple times.
Clicks begin.
Yeah.
I mean,
and there's,
there's a lot of things that I think are well known.
Well,
trodden,
accessibility,
stumbling locks,
things that if you've put time into learning about this,
you know,
but there's a whole frontier of things that just haven't even been discovered yet.
Uh One of my favorite examples of this is like the face id that unlocks your iphone or any other device was not working on dark skinned faces,
right?
It wasn't recognizing that it was a face and I don't imagine that this was done with any kind of mal intent.
It was just that the model was trained on the faces of the developers who built it.
And as it turns out that didn't happen to include any um dark skinned faces.
And,
and so the moral of the story is maybe,
you know,
try for more diversity on your team,
but you're never gonna get it,
achieve that perfectly.
But the the bigger moral is that there's always going to be something that you just did not anticipate.
That is a major mistake,
a major miscue where a whole population is not included in what you build.
And,
and how do you go about addressing that?
Talking about it?
Honestly uh when something like that happens.
So I think that it is,
you know,
I I'm really grateful and,
and have access to a lot of resources working at Google.
We do a lot of research,
we have a lot of best practices.
It doesn't mean we cover all the cases by any means.
But I think that as both individual companies and as an industry sort of sharing these things and not doing what is like just what's legally required by the ad a or other legislation in,
in different places.
Like,
can we share this?
Can we actually say what is best for all users is for us to catalog like users who do not have sight or have,
you know,
low contrast site.
Should we have a mode or should interfaces by default handle that or should we always have audio description for everything and not just rely on their screen reader,
but is their screen reader gonna do a better job?
Do they prefer their screen reader to our audio prompts because it goes at a speed and a cadence that they're used to and they don't want to listen to our sort of sweetly toned voice,
speak very slowly to them when they're trying to get their work done.
Um Having those considerations and empowering users to make choices like,
yes,
we can overwhelm users with choices.
Yes,
we can make it indecipherable.
What all of those choices do?
So that's wrong in a different way.
But can we find a way to say,
what about if all of our products?
What if about at the OS level if you have high contrast?
Can every app respond to that?
And you know,
I think in some cases some of these types of things are supported by Android and I Os and Mac Os and Windows and Linux and things.
But like what are the A as you mentioned,
Brian?
Like what are the other things that we're missing?
And when we discover it,
how does that get shared and how does that get more globally addressed?
Like,
how are we sure that our products aren't just usable but respectful,
like being respectful to a culture has a lot more to do with how people are feeling using the product than if you translated the U I like interesting and like,
yes,
we should translate and also we should use native speakers to validate that translation and not just do word for word translation because we know that that's going to be wrong and we know it's either gonna be offensive or it's going to mean something perverse or whatever,
like whatever it is,
the very least it's going to be confusing.
You know,
we've all,
we've all read instructions,
booklets that were,
you know,
obviously written in another language and then translated word for word into English and they're comically wrong.
So obviously we're doing the same thing going in opposite direction.
Sure.
But if you put a black and white photo of a baby is,
is that harmful?
Like is that how people like capture deceased people in their culture?
And now your product is promoting dead babies?
Like that's not great.
So let's try not to do that?
But how do we find out?
And again,
how do we share this information?
And can we,
and,
and I feel this way about a lot of things that we do as uh you know,
software professionals is like,
have we learned anything like we've been doing this for roughly 60 or 70 years,
this being some form of professional software development,
like building software.
Um and we make a lot of the same mistakes over and over again.
Um,
it's not like people are confused about how to,
can't deliver a bridge and have to like,
do that by guess and check now,
like that has been part of the knowledge that is sort of recorded and passed down through that discipline for thousands of years.
But we are like,
it,
it feels like in our infancy it's like,
ok,
cool.
Well,
you know,
we don't have to,
you know,
use the punch cards and stick them in and keep them in order.
Ha ha ha.
Like we don't have to do that anymore.
It's like,
yeah,
yeah,
but we still have to,
you know,
how many times has security been messed up in the same way?
Like,
how many buffer overflows do we need to have before we stop?
And it's like we haven't found the number because it still happens.
Do you think that part of that is because of the fact that in our industry,
you don't need,
say a doctorate to work?
And so you don't need,
you don't need a certain number of years of apprenticeship and there's not as much oversight.
Exactly.
I mean,
I think that we're a fairly,
still unregulated industry,
at least when it comes to barriers of entry to become an engineer,
you don't necessarily have to have certain degrees.
You don't have to have certifications and I'm not sure,
I wish that we did.
But what we're talking about here is possibly a consequence of that.
Yeah,
absolutely.
If,
if you don't have that,
where does the institutional knowledge come from?
It tends to be the first job people might have either whether they came from university or they came from a boot camp,
the first place that they professionally develop software,
they're going to learn something about what that means.
It doesn't mean it's industry best practices,
although what our industry best practice is even.
But like,
yeah,
we,
we don't,
we don't have a codified way to do that.
And as you mentioned,
like,
we don't have a professional seal that you get when you pass a particular exam and we don't regulate what it means.
And like,
yes other countries in Canada,
you cannot say you're a software engineer if you don't have a pe,
so that's a thing that they enforce.
Ok,
fine.
But we don't hear in the US and a lot of places in the world,
whatever,
you're writing software and how many people could die.
It's lots like we don't like to talk about that very much,
but like the impact of bad software is tremendous.
Um So is it that we don't have apprenticeship?
For example,
if we did have apprenticeship,
what would that mean?
We can't agree on what the best way to build software is or a standard way to build software.
So who are the,
who are the people that they're,
you know,
the journey persons or whomever that these apprentices work under?
And are they just how is that any different than just at their first job?
The engineer that you know,
is checking their code saying,
oh no,
we do four loops like this.
There's a topic that's kind of massive that I would like to get to today because it kind of gets to the core of all of this because all of this is great in theory,
lots of people are well intentioned,
but it still doesn't get done.
And why is that?
All right folks,
that is a big topic and we are gonna have to get into it next week.
Thanks for listening to part one.
We will be right back here next week with part two of our discussion with Lee M keen on empathy and inclusion.
We'll catch you then.
Thanks for listening to the main thread.
As always,
the views expressed by the hosts and guests of the main thread are our own and do not reflect those of our respective places of employment are you enjoying the podcast?
Do you have a topic you'd like to hear discussed in a future episode?
Well,
please reach out to us with any suggestions or feedback.
We are at the main thread on all major social media platforms including Facebook,
Twitter,
Instagram and of course reds if you prefer,
you can send us an email at main thread podcast at gmail dot com.
For links to resources mentioned in today's episode or to download a transcript of the podcast.
Please refer to our show notes that's all for today's episode.
Thanks again for listening and we'll catch you next time on the main thread.