August 12, 2025 • 34 mins
What does it take to land a data analyst job at Tesla, and what challenges await you once you're there? Join me as I interview Lily BL, a former Tesla data analyst, who reveals her exhilarating journey in the world of data at one of the world's most innovative companies.
💌 Join 10k+ aspiring data analysts & get my tips in your inbox weekly 👉 https://www.datacareerjumpstart.com/newsletter
🆘 Feeling stuck in your data journey? Come to my next free "How to Land Your First Data Job" training 👉 https://www.datacareerjumpstart.com/training
👩💻 Want to land a data job in less than 90 days? 👉 https://www.datacareerjumpstart.com/daa
👔 Ace The Interview with Confidence 👉 https://www.datacareerjumpstart.com/interviewsimulator
⌚ TIMESTAMPS
00:00 - Introduction
00:31 - Working on Data Projects at Tesla
01:45 - Was it challenging working at Tesla?
08:34 - Hiring Process and Employee Evaluation
11:56 - Tools and Technologies Used
13:38 - Lily Landing the Job at Tesla
15:42 - Advice for Aspiring Data Professionals
19:36 - How the Data Analytics Accelerator helped Lily
25:11 - Data Analyst Titles Matrix
29:50 - Linking Business Needs to Data Solutions
🔗 CONNECT WITH LILY BL
🤝 LinkedIn: https://www.linkedin.com/in/lilybl/
🔗 CONNECT WITH AVERY
🎥 YouTube Channel: https://www.youtube.com/@averysmith
🤝 LinkedIn: https://www.linkedin.com/in/averyjsmith/
📸 Instagram: https://instagram.com/datacareerjumpstart
🎵 TikTok: https://www.tiktok.com/@verydata
💻 Website: https://www.datacareerjumpstart.com/
Mentioned in this episode:
📰 Join 18k Other Analysts in My Newsletter
Get 1 weekly newsletter helping you take the next steps in your data career! Check it out: https://datacareerjumpstart.com/subscribe
1:1 Coaching w/ Avery
I've opened 12 slots on my calendar to do 1:1 coaching calls. What we can do: fix your resume (it may suck), audit your LinkedIn (also may suck), mock interview you (prepare to sweat), or tell you exactly what to do next This is a test. Never done it. Might never do it again. Limited spots. Book now.
https://landadatajob.com/coachingday
✨ Try Julius!
This episode is brought to you by Julius – your AI data analyst companion. Connect to your database and/or business tools, pull insights in minutes–no coding required. Thanks, Julius, for sponsoring this episode. Try Julius at https://landadatajob.com/Julius-DCP
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Avery Smith (00:00):
This is my Tesla, but I've never worked for Tesla.
Luckily, one of my acceleratorstudents has Lily BL, and today I
had the chance to interview her.
I got to ask her what it was like to workat such a unique, cool company, what it
actually took to get her there, and whatadvice she'd give to those of you watching
who are interested in working at Teslaor other really cool tech companies.
(00:23):
So let's go ahead andget into the episode.
Lily, your career has taken youto Tesla, pg and e, Intel, and
now the city of San Francisco.
But your first full-timedata job was at Tesla.
What was it like workingon data projects at Tesla?
Uh,
Lily BL (00:38):
it was nerve wracking and exhilarating at the same time.
I was not sure what to expect becausewhen I took the role on, it was blended
between a couple of different areas.
But as, uh, I worked more andmore like day after day, I could
see what their data needs were.
They had data in multiple systems formultiple reasons, and it was just so
(01:00):
much in volume that they couldn't keeptrack of how to look at it concisely.
They had to go through embeddedrecords to get an answer.
So I think when I first got fired on,the word on the street was she's like an
admin assistant to the district manager.
If you have administrative stuffyou can't do, just give it to her.
Once the boss saw what I could do.
(01:21):
It completely changed, and Iwas monitoring everything for
data flow to determine whatkind of visuals could be built.
The scary part was like, I don'tknow, and the exhilarating part
was, but I can figure it out.
Avery Smith (01:33):
That's awesome.
And I'm glad to hear that.
Uh, an employer like that was, youknow, first off recognized your talents,
but then second off was like, okay,Lily, Lily can, uh, do this stuff.
Let's give her some, some more tasks.
Did you feel like what, what you weredoing, like was, was super cutting
edge or did you feel like it wasmore like regular, regular tasks?
(01:54):
Like, um, did you feel likechallenged in what you were doing?
Lily BL (01:57):
I did feel challenged in what I was doing because it had a
lot of impact once it was completed.
The technology and the how to itselfsurprisingly was very basic, so it was
continuously searching for the specificthing that will fix this specific problem.
And then gathering all the solutionsto say, Hey, this is how you can
(02:17):
improve your data situation overall.
Avery Smith (02:21):
That's interesting.
And, and correct me if I'mwrong, uh, you know, Tesla is
obviously a, a large company.
I worked for ExxonMobil, a largecompany at these large companies.
You hear this phrase, I'd neverreally heard it before, this word.
Um, it's called disparate.
Uh, and basically, or siloed, Ithink is the other thing that they
said a lot at Exxon that data issiloed or that data is disparate.
Basically, I think you kind of saidsomething similar where, you know,
(02:44):
at these large companies there's alot of data, um, and there's a lot of
systems, but the problem is this systemdoesn't necessarily talk to this system
and this data is kind of stuck here.
And you know, this data, they onlyenter it in an Excel, uh, database.
So it doesn't really likeintegrate with anything else.
And so it sounds like your job was almostlike you were like data analyst glue.
(03:04):
To try to tie in all these differentdata sets from this different systems.
Did I, did I get that wrong, oris that kind of what you did?
Lily BL (03:10):
Yeah, you're nailing it on the head.
It was very interesting becausea large portion of the actual
engineering work was done insideof a software called Jira, which is
meant for tracking, uh, the project.
That's the data that was neededto be reviewed and the company's
decision was not to view it.
Outside of that, I would run validationsin Excel to make sure the numbers it gave
(03:33):
me were accurate to what I visualized,and so I had to actually learn Jake to
be able to put together what I neededand I was limited by the visualizations
preselected for project management.
At the end of it, it was super coolbecause I kind of created a grid.
Um, it was like a large,uh, standing rectangle.
(03:53):
When you looked at it up and down.
That was the information for the district.
Managers.
Managers.
So each team and all of their staff.
You could look at what theydid and when they did it, what
was still pending Vertically.
Each team had a, a call.
When you looked at it horizontally,those were all the KPIs my
district manager had requested.
(04:15):
So they were subject to some standardsand it was so cool because I didn't know
how to do that till I was done with it.
Um, and so then I was like,yeah, this is what I wanted.
Uh, but I needed some help fromthe management team because to
make sure the data had integrity,I didn't set the conditions.
So I did push for them to tellme, this is how this is defined.
(04:36):
This is the thresholdto, uh, determine this.
And once I had them deliver to me,uh, some definitions, I used those
decisions to build out the, thevisual and they ended up loving it.
'cause I was able to color codeit and I assimilated it to red
light, green light, yellow light.
So if it's marked green,don't worry about it.
You don't have to look.
(04:56):
If it's yellow, you kind ofneed to keep an eye on it.
But if it's red, you needto go in and investigate.
That was just relative to thedata produced by the teams.
Then there was the data producedby the hardware and that
those were different systems.
So then in those systems, I tookthat one to Excel and I was able to
create a chart that had a threshold.
(05:17):
So I had them, again, definewhat the threshold was, and
let's say it was like 10.
Once you got 10 of these things,the chart would go from being green.
To now being red 'causeit crossed the threshold.
So the division manager could lookat all of these things at a glance
and be like, oh, red is where I needto be, and figure out what happened.
Um, and then, uh, separate from that, Iwas also, uh, building into Tableau, uh,
(05:42):
master portfolio so that the divisionmanager could just look in there at
all of the teams and all of the thingsthat were of interest to him because he
would take that information back to themeetings with the rest of management.
They would decide what would comenext based on what was there.
So like if a lot of equipment wasfailing, they would say, Hey, your
team is under producing because youhave 10 of these different kinds of
(06:05):
machines and you're only putting outlike half of the results we expected.
Or based on the analysis, half ofthe machines were not available
for one reason or another.
So it's like, this is why our numbersare lower than what's expected.
Per what is available.
Only half is available, which youcouldn't really tell any other
way is they were kind of, uh.
Digging constantly before we wereable to build the visuals, uh, to
(06:27):
determine, uh, what was really going on.
So it really facilitatedthe manager to manage.
And he was actually a really good manager,so he knew where the weak points were.
He just was not a data person or a dataanalyst to be like, I need this, felt
like this, and like that to get this.
But he knew what he wanted, so itwas a perfect partnership because
I could build it and he could tellme if it worked or didn't work.
Avery Smith (06:48):
Lily, this is super cool.
Thanks for sharing all of this.
Um, I have so many places I, I wantto go based off what you just told me.
Uh, the first was, I had neverheard of J Quill, but I had
a chance to look it up here.
So that's, that's Jira QueryLanguage or Jira, I don't know
how to say that, but for those whonever heard of that, it's JIRA.
Um, and it is owned byAtlassian, I'm pretty sure.
(07:09):
Um, and it is like a project managementsoftware that a lot of, uh, software
companies use to develop software.
So you're, you were kind oflooking at project data, um,
which is, which is really neat.
Uh, and it sounds like you were workinglike really close to these, you know,
kind of higher up stakeholders who,you know, they need a bird's eye view
(07:29):
of what's going on in their business.
It's, they kind of, like you said, havelike a gut feeling of this is, you know,
maybe this part is struggling righthere and I have a feeling why, but I'm
not exactly a hundred percent sure.
What it sounds like is you tied up abunch of loose ends and you know, this,
these disparate data sets, and you'reable to create a data visualization, uh,
(07:50):
that helps these managers see, you know,maybe what's struggling in the business,
maybe what's doing good, um, what theyneed to worry about and what they need
to maybe put their, their focus on.
Um, so it sounds like you were almostgiving them like supervision goggles to
like look into their business and likeactually see is everything, is everything
going the way it's supposed to go because.
(08:10):
You know, as someone who runs abusiness, I obviously do not run a
business close to the scale of Tesla.
Like one division of Tesla, I'm sure isa hundred times bigger than my business.
Uh, but even now I have, you know, TrevorMaxwell helping me out with coaching.
I have Isaac Ania who'shelping with my community.
I have, uh, a podcast producerand editor, and I don't know
what's going on half the time.
(08:31):
They're just awesome.
Uh, employees doing a great job,but I do wanna be like, okay.
How do I get above the business and likeactually look down on it and see like,
okay, what's going well and what's not.
And it sounds like you were ableto do a bunch of analysis to kind
of produce that for these managers.
Lily BL (08:46):
Yeah.
And uh, one project, uh, that I was athair away from completing, 'cause I was
missing one definition, uh, which I think,uh, would have had a huge impact, is,
um, I worked on their hiring process.
So I would sit in on theirmeetings and see how they went
through their hiring processes andwould sit in on the interviews.
And then I would also look at fubu.
(09:07):
They had already hired because of theway they, uh, did bonuses there, the
managers had to divide a percentage ofbonus among all of the existing teammates.
So I was able, based on watchingthe data flow, um, I was able to
determine what the standards were.
They already knew that they had tiers.
Like we have engineer 1, 2, 3, 4,5, a lead, whatever, a, a manager.
(09:30):
Then I was able to zero in on whatare the standards, uh, that this
person needs to complete or beingknowledgeable in, in order to ascend
to the following tier, which translatesto more money for the employee.
And so then, um, we got, I got, Ireviewed everything and I set it
up, but what was missing was themetrics associated to each tier.
(09:52):
And so I left it aloneto not push a project.
Uh, and hack the pay be incorrect.
At the very end of the, uh, contract,the HR published the standards
or the, the pay for each scale.
So that was the missing piece I needed.
But with that, it would've facilitatedall of the yearly reviews of the
(10:12):
management team to enable or todetermine very quickly, oh, this person
hit these projects and these projectsare labeled within this category.
So while they were working asan engineer too, their work.
Function was actually anengineer, four or five.
So they qualified for the bonusand potentially a promotion.
Um, we were also very proactive there withkind of working with the, um, employees
(10:37):
being more, um, uh, how do you call it?
Um.
It's not affirmative, but it's beingmore proactive about their findings
and stressing their good works.
So with so many people on the team, Idon't know every single thing you did,
so take the initiative and tell me,Hey, I completed these multiple things.
That way it's fresh on my mind.
(10:59):
So I didn't get to the see there.
But after I would have completedthat project with the raids, my goal
would've been to work with the teamone-on-one and have them pitch me.
Their successes and then Icould categorize it for them.
Like, okay, what you saidfalls into this or into that.
Do you agree or disagree?
And then teach them how to make theargument for their good works better.
(11:20):
Um, it's delicate to do in business,but it's like a negotiation.
So you actually need topractice it in order to get it.
And this particular company wasopen to that they wanted to hear.
So, uh, that was like theicing on the cake for me.
We didn't get to finish it, butit's something that would've
been proactive for everybody.
The company would've had a very,well, a very articulate staff, which
(11:43):
is needed for problem resolution.
And then with the market as it wasconstantly laying off and whatnot,
this employee would have had theskills sharpened to then go right
into another position if they werelaid off and quickly get another role.
Avery Smith (11:56):
Hmm.
Very cool.
Um, while you're at Tesla, whattools did you use the most?
Lily BL (12:01):
Um, I think I used Jira the most and excel for the validation.
Um, I got heavier into theadministrative side of, of the
software because for Tableau and JiraI was bringing in add-ins to make
them more functional for analytics.
Uh, so for companies, uh, you have toconnect the Tableau software to what,
(12:22):
wherever your data is in the company.
When you use Tableau as anindividual user, you just
connect it to your worksheet.
You can't connect it to somethingelse if you have it, but typically
you just use a worksheet.
So that was different.
And it was a full hostof security clearances.
Um, so I did a little bit of theadministrator stuff, but, uh, JIRA
and Excel round my validations.
Avery Smith (12:43):
That's awesome.
I think that's true.
And, and you mentioned JQL.
Is that kind of like SQL orhow, how are those related?
Lily BL (12:49):
Yeah, so it's very similar to the commands in sql, so that's why I
was able to learn it pretty quickly.
Uh, but then, um, somethings are specific.
It uses a lot of, uh, a lotmore keywords than you would
expect, and they're different.
Um, the software itselfdoes try to help you.
Like it lets you click on buttonsand produces the code for you.
(13:10):
Uh, to an extent, but then youhave to have modifications.
So I would allow the software to allowme to click to build some of the stuff,
but then I would review it and determine,oh, it still needs this functionality,
or this, or this other group of people.
And you would have to manually put thatinto the existing code to make it function
Avery Smith (13:28):
super neat.
So it's basically SQL for Jira, andthey try to make it a little bit easier
for you to actually write the code.
Um, okay.
I'm actually not sure theanswer to this question.
How did you get this job at Tesla?
I remember you messaging me whenyou got the job offer and you're
like, Hey, these are the details.
What do you think?
Should I take this job or not?
You know, it's one of the things I tryto do with my accelerator students, but
(13:49):
I don't remember off the top of my head.
This was a couple years ago now.
Um, how you ended up landing this job?
Lily BL (13:54):
Yeah, I think it was through networking.
Um, at the time I was an instructorfor co-op and I had a cohort that
I would teach in the evenings.
One of my cohort students actuallygot hired by them about a month or
so before I helped them finalize hiswork that they wanted him to see.
So then about a month or so later, I got,uh, contracted by a recruiter on LinkedIn.
(14:16):
I checked with him.
It ended up being the sameperson that contracted him.
So I think what happened is that Ipopped up for her in association to him.
But she never said that.
But that's, that's what wethink the connection was.
And so then she interviewed me.
I actually was like number three orfour, because three or four other
people had said yes and then backed out.
(14:39):
And so then it was really easyto consider, oh, you know what?
I just won't take the job.
You know, it just seems really hard.
But I just kept saying, well,if the, if the manager wants to
interview me, I'll be available.
If the position comes back open.
So after that is how everybodyelse, 'cause there was a lot
of things that it required.
And then I ended up notdoing most of those things.
(14:59):
Uh, so I ended up, uh,hanging in for the interview.
And then in the, in the interview,um, he asked me some questions that
I think everybody else struggledwith and I answered very confidently,
uh, because of the work that Ihad done inside of your bootcamp.
Avery Smith (15:13):
That's, that's awesome to hear.
So, uh, what I was kind of hearing was.
Basically you were, you were connectedto the, to a right person, someone
kind of similar, one of your peers,um, looking to inundate a job.
Uh, and then you had a goodLinkedIn because the other thing
is, uh, on, on LinkedIn, right?
Like you don't get recommended ifyou have kind of a crappy LinkedIn.
So making sure your LinkedIn wasup to date with all the right
(15:34):
keywords, all those projects youhad done inside the accelerator,
I'm sure that helped, uh, as well.
Yeah.
You, you nailed the interview.
Okay, that makes sense.
So what advice would you give to someone.
Who's listening right now who's like, wow,I wanna be cool like Lilly and work for a
cool company like Tesla in the data space.
Like what advice would you give them?
Lily BL (15:52):
Um, I would recommend that they kind of determine what part of the data
portions they like to do, and then afterthey figure out, oh, I like building
the data structures, or the pipelineor the visualizations, dive into that.
I do get, uh, a lot of requests forlike, how can I kick off figuring
out my data stuff and actuallyrecommend them to your free content?
(16:13):
'cause I find it really helpful.
I think you do a good job organizing.
Okay.
You gotta go get the data.
Once you get the data,you gotta clean the data.
Well, once you clean the data, you gottafigure out a quick way to deliver it.
And also the visuals, how youbuild your visuals is gonna kind of
determine what you can say, um, asthe Bluff Fund, like right up front.
I know these are things we do to buildthe projects, but those directly translate
(16:36):
into the interview and also into workingwith, uh, people on site in the jobs.
So if you can find a material that helpsyou hone the skills you naturally want or
like inside of your data, uh, career, itwill make it easier for you to get that
and it will make you a natural to post it.
'cause you'll actuallybe excited about it.
(16:56):
Like, oh, I had a hard timelearning this particular function
in Excel, but I nailed it.
Let me show you guys how I did it.
You'll naturally be like, oh, wehad overtime with sql, but I figured
this out and now I'm gonna post it.
And people actually do look at it.
They might not comment, they mightnot like, but recruiters and also
other people, uh, interested in datawill come and look at your projects
(17:18):
because if you had an issue withit, likely someone else did too.
So if you're constantly posting yourprojects and how you solve the problem.
Uh, they will naturally gravitate to you.
And one thing I always stressis to try to frame my projects
into a problem and a solution.
So the purpose of this projectwas to address this specific
problem and here's the solution.
(17:40):
Maybe they won't care for lookingat the problem, but maybe they're
interested in just a solution.
But that's interesting.
They'll go back and look at theproblem and then read all of the work.
Avery Smith (17:48):
Interesting.
So yeah, projects played a bigrole for you, it sounds like, like
you really believe in, in doingprojects and then posting them on
places like LinkedIn to get noticed.
Lily BL (17:57):
Yeah.
And in the interview for Teslaspecifically, uh, I think the
question that sealed the deal forme was that, uh, the bus had asked
me what I would do in inside of sql.
So he asked me just the generalstuff, like, you know, how would
you get something to come up?
What would you call the tables?
And he goes, it was likehis, his secret question.
It was supposed to catch me off guard.
He says, what if thereisn't anything in there?
(18:20):
Like you asked for it and it doesn'tgive, like there's nothing in there.
What seat we're going to do then.
And we, I had done a module, uh,to the bootcamp, uh, that you have.
And I had picked the short dataset instead of the large data set.
And because I picked the short dataset, my results were different.
And in fact, missing.
Mm.
So I distinctly remember sittingthere for like, what, what happened?
(18:42):
Did I do it wrong?
Rewatching the video, redoingthe thing, and trying and trying
until I got very frustrated.
And then I realized, oh, I pickeda different data set than he did.
So our results are probably not the same.
They're likely missing from mine.
So then I manually went in and checked,and that was exactly what happened.
So when this manager from Tesla asked me.
I knew exactly whathappened when that occurs.
(19:03):
And so I was like, youget absolutely nothing.
It's the most frustratingthing in the world.
Uh, but it's good because youdon't have to keep looking.
There's absolutely nothing there.
You, you're just gonna get a, andbecause I was so confident about it,
having sat in the frustration, helaughed and then was like, I, I think
that she will be able to figure outwhatever she doesn't know and what
she does know will benefit us anyways.
(19:26):
And I think that's what sealed the deal.
So as you're working through theprojects and honing your skills.
Think about what you experience.
'cause that's what's gonna makeyou shine in the interviews.
Avery Smith (19:35):
I, I love hearing that.
I love hearing that the experiencesyou had inside the accelerator
program, uh, worked out wellfor you in, in an interview.
And it's interesting because, uh, Iobviously try to design the accelerator
and we're constantly updating it so thatpeople have less and less problems, right?
Like, we wanna try to make it as easyfor people to learn data as possible.
But the silver lining iswhen those problems happen.
(19:56):
It puts you in a real lifescenario 'cause you're gonna have
problems when you get on the job.
And figuring out how to solvethose, figuring out what's
going wrong, uh, is a skill.
It's kind of a hard skill to teach.
But it's a very valuable skill to have.
So, uh, I love hearing that,you know, a lot of data skills.
I'm curious here what order youlearn them in, and if you have any
(20:17):
tips for anyone who is learningthese different data skills?
Because there's a lot, right?
There's Power bi, there's Tableau,there's Excel, there's sql, there's
Python, there's R Like what order didyou learn those in, and what advice would
you give to someone else learning those?
Sure.
Lily BL (20:28):
Uh, I think the order I learned them in was first Excel
and the Microsoft Office Suite.
Uh, I actually was certified throughthem, um, to use Word in Excel.
However, I didn't understand itas much as I would over time.
So then with the Excel basic knowledge,I was able to navigate most data
and then I realized everything'strickling into information systems.
(20:51):
So when I realized that I went back toschool and I got a degree that focused
in information systems and there I wasintroduced to, uh, data visualizations
where we used a variety of tools.
Uh, the one that stood outthe most to me was Tableau.
So from there I joined an apprenticeshipwhere they used that tool.
(21:12):
'cause it just was visually stunning.
The rest of the stuff could getthe things done, including Excel,
but they were kind of grainy.
But with Tableau.
You could just floor somebodyby just the visual alone.
You wouldn't have to say anything.
They'd just be looking at it for a while.
So I was like, I'm reallyinterested in that.
So I did that.
And while I was in that program,we also covered, uh, Python,
(21:34):
uh, more basics and sql.
And, uh, we also did, uh,presentations, um, of the findings.
After I had, uh, those things under mybelt, I discovered your bootcamps and
then went back to square one with Excel.
Was like, okay, this is how you useExcel specifically for data analysis,
not the other stuff I was doing.
(21:55):
So it redefined, like, it reallysharpened what I knew how to do.
And from there, uh, I went back into sql.
A lot of the companies I worked for didn'tuse SQL as intensively as I expected.
So I was more so, uh,using Tableau frequently.
And then Power bi.
Uh, power bi, um, islike a full stop shop.
(22:16):
For analytics because it allowsyou to do the visual component.
But to do that you need tobe able to pull in data.
To pull in the data, you needto understand like the, uh, the
stakeholder request, and then also howto clean the data and it uses Excel.
So, um, the skills were the samein all of the software you just
(22:36):
clicked in a different spot.
So throughout the software per uh.
Phases or processes.
What I was continuouslysharpening was what is the data
process independent of the tool.
So if I had to start all the way over,the way that I would learn these in
is Excel, uh, power bi and then uh,Tableau and last sql, unless it you
(23:02):
are company that you're targeting does,is focused on sql, I would do Excel
and then sql because if you understandwhat you're doing in Excel, like, um.
V lookup, a next lookup, an H lookup.
They're essentially joining data.
So if you know how to join the data inExcel and you can articulate it, then
you can look at any other software.
Here's um, a sql, let me goahead and join data here.
(23:26):
This is how I do thejoins in this software.
Okay, now I have Tableau,how do I do the joins here?
And you are specifically honing yourskill for joining data, which is like
the backbone for, uh, data analytics.
And then that will parlay youinto engineering if you want.
Uh, but I would go Excel first andthen whatever you learn in Excel,
(23:48):
mirror it in whatever softwareyou can get your hands on next.
I did have the cases sometimeswhere I didn't have certain
software, so I've had to wing it.
Um, I did a lot of G docs andthe, all of the Gmail suite
documentation when for some time Icouldn't afford the office software.
So even if you can't getthe most premium thing.
Do what is affordable, but focus onthe skill you're trying to sharpen
(24:12):
and you'll be able to figure it outeven if you've never used it before.
I
Avery Smith (24:15):
think that's a really cool, uh, point there is like, you know, we use
different software at different times,but really a lot of them do similar stuff.
They get data from places.
You clean data with them, you do somesort of aggregations or analysis or
make some charts obviously, like SQLdoesn't really make a lot of charts.
Like a pivot table in Excel is prettymuch just like a group buy in sql.
(24:38):
Um, so there is a lotof, uh, overlap there.
So that makes a lot of sense.
So Lily, when you were tryingto break into data, there's
obviously a lot of data roles.
Um, there's data analysts,there's business analysts, there's
operations research, which iswhat I used to do at ExxonMobil.
Um, and each one of those jobs,uh, is kind of complicated.
(24:59):
They, they're all data analyst roles, but.
They have different domains,they have different industries,
they have different focuses.
They may use different tools, they mighthave different vocab and, and customers.
So one of the things I really love, um,that, uh, you sent me was like this matrix
you made of a couple different, uh, dataanalyst titles and what you'd be doing
(25:22):
slash what tools you'd be using basedoff of how experienced you you were.
So tell me about this matrix you made.
Why did you make it and, youknow, what does it do for you?
Lily BL (25:32):
So I wanted to share this, uh, with you and, uh, potentially
to anybody trying to break into dataor further career in data, because
this is how I was able to do it.
Uh, pretty much when you startat the beginning, you don't
have a bunch of experience.
Um, in my case, I just knew Excel,but not specific to analytics.
So the way that you leverage, uh, thetool I gave you is that you kind of.
(25:55):
Set up your goals by a five year plan.
And the reason why is because by thefifth year of any profession, you're
considered a professional 'causeyou've been in it for five years, you
have enough working hours to do this.
At a professional level,you're not guessing anymore.
You should know, uh,concretely what you're doing.
So, uh, depending on what kindof analytics you wanna do,
(26:16):
the matrix can kind of guideyou to where you would start.
Let's use me for an example.
I started with Excel and Iwanted to be a data analyst.
My first data rules werenot titled Under Data.
So what I did is that I said,Hey boss, I know you want me to
take care of these appointments.
And it was clerical work, but itwas, uh, handling a lot of data.
(26:36):
So I said, Hey, you have anopportunity here, uh, to figure out
why your patients are dwindling.
So I took it upon myself tooffer a project so that I
can gain the skills I needed.
So in that project I recovered about halfa million dollars, uh, of lost payments
because somebody clicked the wrong button.
And there I secured my Excel experience.
(26:57):
I secured, uh, the patients beingable to return the company, gaining
the money, uh, that had originallybeen lost because, uh, I used Excel.
That's what I needed in order to beginto say, Hey, I have six months work.
With Excel, I have ayear's worth with office.
Um, at the time it was very popularto use the Microsoft Office Suite.
Uh, let's say you secure thethe time you need with Excel.
(27:20):
Now you can say, Hey, in Excel.
I've also executed Pivot chartsand VLOOKUPs so I can join data.
I'm ready to go onto the next thing.
Hey, boss.
Uh.
We have a lot of data ina lot of different places.
We already are integrated with Microsoft,so we can use Power BI to pulling
all the data sets into one location.
(27:41):
Uh, can I get some time to be ableto make that happen so that I can get
you some support with your recordingand then you start figuring that out?
You might, when you, when you do this,you don't have necessarily somebody
coaching you, so you need to relyon the bootcamps or the knowledge
you already have that gives you theconfidence that I can execute this.
If you can't execute in thesoftware you're reaching for, don't
(28:03):
nominate yourself to do the project.
In there, you do it 'cause youalready know you have, you know
how to use that software, but thecompany's just not implementing it.
So then you would jump into Power bi.
Maybe not its most advanced things,but just enough to get your feet wet
so that you can figure out, this ishow I use it, this is what I like.
Once you get in there,you can be like, Hey boss.
Uh.
We're in here with the Power bi.
(28:24):
We have these simple reports, but we havea lot of stuff inside of SQL as well.
I was wondering if you can get me access,uh, to request permission to join them
into the Power bi and that way I canaccess more data and goes from there.
Right.
Well, one of the visualizations inPower BI is a table, so you can actually
organize and clean all of your data insideof Power BI and then export that sheet.
(28:49):
Put it into somethingstunning like tablet like.
It's hard because as you're workingon it, it's not inherently clear
what you're doing, but that's howyou use the document I sent you.
You look at the title you want,what software or what knowledge do
I have now and what can I reach forbased on my hidden skills that I
can start to attribute to my career?
(29:09):
And that's why you slowly grow it.
Now, sometimes the companies will say,no, we don't need any work in Power bi.
We just want it done in Excel.
For me that translated to, I need tofind another company because I really
wanna grow more skills, uh, to get tothe next level because I have five years
to make it to that professional status.
And if I hit five years and I don'thave all the things in my tool belt,
(29:31):
I gotta do more than five years.
That's what I used toget into the next thing.
Um, also, if you don't wanna growyour career, like you're happy with
what you're doing, don't volunteerthe projects or the software.
Um, hone on what you, or focus yourskills on honing what you already
know and that will make you sharperand sharper with what you have.
Avery Smith (29:50):
Well, I think that's one of your skills is that you're really good at
figuring out how to link business to data.
Uh, and I think a lot ofbusiness and operations people
kind of struggle with that.
Um, so it's really cool that youwere able to be like, Hey, I see this
business need, uh, here's how analyticscould help us, uh, in this case.
Um, and I think that, you know,you've done that as well with building
(30:13):
dashboards for, for stakeholders thataren't necessarily, uh, data experts.
Um, I guess how do you have like an eyefor where data can help these businesses
and how do you, uh, help these maybenon-technical, non-data folks be excited
and interested and ready to, to helpwith these data analytics projects?
Lily BL (30:36):
Well, that's such a, that's such a good question.
Um, because you kind ofhave to actively listen.
So, uh, it's almost likespeaking another language.
Somebody can say, oh man, like in real,a real life example, a boss that I had
said, oh, I just want this inside ofExcel, and I'll be happy if we could
just get it from where it is into Excelso that I can analyze it, I'll be happy.
(30:59):
So I got it done.
After it was done, they were sohappy that they decided, I want,
I wish everything can go in there.
And I said, what?
What you want, sir?
Is a warehouse of data.
Mm.
So what he said is, I want everythingin there, or I want this in Excel.
But what they're asking foris an accumulation of data.
They're asking for a pipeline.
If you understand the data portion ofthat, you can translate the regular
(31:23):
English into what that looks like in data.
And that's how you can determineI can fix that or I can give you
something to help you hit that goal.
Or you can determine, oh, you knowwhat, that's just outside of my reach.
'cause your Google and you have a bunchof data, I can't handle that much stuff.
Like I need servers, I need abunch of other stuff, but these
portions I can handle for you.
(31:43):
And that's how you determine I cando this versus I can't do that.
I should offer you this 'causeI know I can execute that.
Avery Smith (31:50):
Lily, that is awesome.
I think that is a superpowerthat, that you have.
Thank you so much for giving a glimpseinto what your career was like, telling
us what it was like to work as a dataanalyst at Tesla and give us some
good, uh, advice and feedback for.
Trying to learn these data skills andtrying to maneuver in our data careers.
Is it okay if we put your, uh, LinkedInin the show notes down below and if people
(32:12):
have questions they can reach out to you?
Sure.
Okay.
Awesome.
Lily, thank you so muchfor coming on the podcast.
It's so good to have youand, uh, good to catch up.
Lily BL (32:20):
Thank you.
Likewise.
This is my Tesla, but I've never worked for Tesla.
Luckily, one of my acceleratorstudents has Lily BL, and today I
had the chance to interview her.
I got to ask her what it was like to workat such a unique, cool company, what it
actually took to get her there, and whatadvice she'd give to those of you watching
who are interested in working at Teslaor other really cool tech companies.
(00:23):
So let's go ahead andget into the episode.
Lily, your career has taken youto Tesla, pg and e, Intel, and
now the city of San Francisco.
But your first full-timedata job was at Tesla.
What was it like workingon data projects at Tesla?
Uh,
Lily BL (00:38):
it was nerve wracking and exhilarating at the same time.
I was not sure what to expect becausewhen I took the role on, it was blended
between a couple of different areas.
But as, uh, I worked more andmore like day after day, I could
see what their data needs were.
They had data in multiple systems formultiple reasons, and it was just so
(01:00):
much in volume that they couldn't keeptrack of how to look at it concisely.
They had to go through embeddedrecords to get an answer.
So I think when I first got fired on,the word on the street was she's like an
admin assistant to the district manager.
If you have administrative stuffyou can't do, just give it to her.
Once the boss saw what I could do.
(01:21):
It completely changed, and Iwas monitoring everything for
data flow to determine whatkind of visuals could be built.
The scary part was like, I don'tknow, and the exhilarating part
was, but I can figure it out.
Avery Smith (01:33):
That's awesome.
And I'm glad to hear that.
Uh, an employer like that was, youknow, first off recognized your talents,
but then second off was like, okay,Lily, Lily can, uh, do this stuff.
Let's give her some, some more tasks.
Did you feel like what, what you weredoing, like was, was super cutting
edge or did you feel like it wasmore like regular, regular tasks?
(01:54):
Like, um, did you feel likechallenged in what you were doing?
Lily BL (01:57):
I did feel challenged in what I was doing because it had a
lot of impact once it was completed.
The technology and the how to itselfsurprisingly was very basic, so it was
continuously searching for the specificthing that will fix this specific problem.
And then gathering all the solutionsto say, Hey, this is how you can
(02:17):
improve your data situation overall.
Avery Smith (02:21):
That's interesting.
And, and correct me if I'mwrong, uh, you know, Tesla is
obviously a, a large company.
I worked for ExxonMobil, a largecompany at these large companies.
You hear this phrase, I'd neverreally heard it before, this word.
Um, it's called disparate.
Uh, and basically, or siloed, Ithink is the other thing that they
said a lot at Exxon that data issiloed or that data is disparate.
Basically, I think you kind of saidsomething similar where, you know,
(02:44):
at these large companies there's alot of data, um, and there's a lot of
systems, but the problem is this systemdoesn't necessarily talk to this system
and this data is kind of stuck here.
And you know, this data, they onlyenter it in an Excel, uh, database.
So it doesn't really likeintegrate with anything else.
And so it sounds like your job was almostlike you were like data analyst glue.
(03:04):
To try to tie in all these differentdata sets from this different systems.
Did I, did I get that wrong, oris that kind of what you did?
Lily BL (03:10):
Yeah, you're nailing it on the head.
It was very interesting becausea large portion of the actual
engineering work was done insideof a software called Jira, which is
meant for tracking, uh, the project.
That's the data that was neededto be reviewed and the company's
decision was not to view it.
Outside of that, I would run validationsin Excel to make sure the numbers it gave
(03:33):
me were accurate to what I visualized,and so I had to actually learn Jake to
be able to put together what I neededand I was limited by the visualizations
preselected for project management.
At the end of it, it was super coolbecause I kind of created a grid.
Um, it was like a large,uh, standing rectangle.
(03:53):
When you looked at it up and down.
That was the information for the district.
Managers.
Managers.
So each team and all of their staff.
You could look at what theydid and when they did it, what
was still pending Vertically.
Each team had a, a call.
When you looked at it horizontally,those were all the KPIs my
district manager had requested.
(04:15):
So they were subject to some standardsand it was so cool because I didn't know
how to do that till I was done with it.
Um, and so then I was like,yeah, this is what I wanted.
Uh, but I needed some help fromthe management team because to
make sure the data had integrity,I didn't set the conditions.
So I did push for them to tellme, this is how this is defined.
(04:36):
This is the thresholdto, uh, determine this.
And once I had them deliver to me,uh, some definitions, I used those
decisions to build out the, thevisual and they ended up loving it.
'cause I was able to color codeit and I assimilated it to red
light, green light, yellow light.
So if it's marked green,don't worry about it.
You don't have to look.
(04:56):
If it's yellow, you kind ofneed to keep an eye on it.
But if it's red, you needto go in and investigate.
That was just relative to thedata produced by the teams.
Then there was the data producedby the hardware and that
those were different systems.
So then in those systems, I tookthat one to Excel and I was able to
create a chart that had a threshold.
(05:17):
So I had them, again, definewhat the threshold was, and
let's say it was like 10.
Once you got 10 of these things,the chart would go from being green.
To now being red 'causeit crossed the threshold.
So the division manager could lookat all of these things at a glance
and be like, oh, red is where I needto be, and figure out what happened.
Um, and then, uh, separate from that, Iwas also, uh, building into Tableau, uh,
(05:42):
master portfolio so that the divisionmanager could just look in there at
all of the teams and all of the thingsthat were of interest to him because he
would take that information back to themeetings with the rest of management.
They would decide what would comenext based on what was there.
So like if a lot of equipment wasfailing, they would say, Hey, your
team is under producing because youhave 10 of these different kinds of
(06:05):
machines and you're only putting outlike half of the results we expected.
Or based on the analysis, half ofthe machines were not available
for one reason or another.
So it's like, this is why our numbersare lower than what's expected.
Per what is available.
Only half is available, which youcouldn't really tell any other
way is they were kind of, uh.
Digging constantly before we wereable to build the visuals, uh, to
(06:27):
determine, uh, what was really going on.
So it really facilitatedthe manager to manage.
And he was actually a really good manager,so he knew where the weak points were.
He just was not a data person or a dataanalyst to be like, I need this, felt
like this, and like that to get this.
But he knew what he wanted, so itwas a perfect partnership because
I could build it and he could tellme if it worked or didn't work.
Avery Smith (06:48):
Lily, this is super cool.
Thanks for sharing all of this.
Um, I have so many places I, I wantto go based off what you just told me.
Uh, the first was, I had neverheard of J Quill, but I had
a chance to look it up here.
So that's, that's Jira QueryLanguage or Jira, I don't know
how to say that, but for those whonever heard of that, it's JIRA.
Um, and it is owned byAtlassian, I'm pretty sure.
(07:09):
Um, and it is like a project managementsoftware that a lot of, uh, software
companies use to develop software.
So you're, you were kind oflooking at project data, um,
which is, which is really neat.
Uh, and it sounds like you were workinglike really close to these, you know,
kind of higher up stakeholders who,you know, they need a bird's eye view
(07:29):
of what's going on in their business.
It's, they kind of, like you said, havelike a gut feeling of this is, you know,
maybe this part is struggling righthere and I have a feeling why, but I'm
not exactly a hundred percent sure.
What it sounds like is you tied up abunch of loose ends and you know, this,
these disparate data sets, and you'reable to create a data visualization, uh,
(07:50):
that helps these managers see, you know,maybe what's struggling in the business,
maybe what's doing good, um, what theyneed to worry about and what they need
to maybe put their, their focus on.
Um, so it sounds like you were almostgiving them like supervision goggles to
like look into their business and likeactually see is everything, is everything
going the way it's supposed to go because.
(08:10):
You know, as someone who runs abusiness, I obviously do not run a
business close to the scale of Tesla.
Like one division of Tesla, I'm sure isa hundred times bigger than my business.
Uh, but even now I have, you know, TrevorMaxwell helping me out with coaching.
I have Isaac Ania who'shelping with my community.
I have, uh, a podcast producerand editor, and I don't know
what's going on half the time.
(08:31):
They're just awesome.
Uh, employees doing a great job,but I do wanna be like, okay.
How do I get above the business and likeactually look down on it and see like,
okay, what's going well and what's not.
And it sounds like you were ableto do a bunch of analysis to kind
of produce that for these managers.
Lily BL (08:46):
Yeah.
And uh, one project, uh, that I was athair away from completing, 'cause I was
missing one definition, uh, which I think,uh, would have had a huge impact, is,
um, I worked on their hiring process.
So I would sit in on theirmeetings and see how they went
through their hiring processes andwould sit in on the interviews.
And then I would also look at fubu.
(09:07):
They had already hired because of theway they, uh, did bonuses there, the
managers had to divide a percentage ofbonus among all of the existing teammates.
So I was able, based on watchingthe data flow, um, I was able to
determine what the standards were.
They already knew that they had tiers.
Like we have engineer 1, 2, 3, 4,5, a lead, whatever, a, a manager.
(09:30):
Then I was able to zero in on whatare the standards, uh, that this
person needs to complete or beingknowledgeable in, in order to ascend
to the following tier, which translatesto more money for the employee.
And so then, um, we got, I got, Ireviewed everything and I set it
up, but what was missing was themetrics associated to each tier.
(09:52):
And so I left it aloneto not push a project.
Uh, and hack the pay be incorrect.
At the very end of the, uh, contract,the HR published the standards
or the, the pay for each scale.
So that was the missing piece I needed.
But with that, it would've facilitatedall of the yearly reviews of the
(10:12):
management team to enable or todetermine very quickly, oh, this person
hit these projects and these projectsare labeled within this category.
So while they were working asan engineer too, their work.
Function was actually anengineer, four or five.
So they qualified for the bonusand potentially a promotion.
Um, we were also very proactive there withkind of working with the, um, employees
(10:37):
being more, um, uh, how do you call it?
Um.
It's not affirmative, but it's beingmore proactive about their findings
and stressing their good works.
So with so many people on the team, Idon't know every single thing you did,
so take the initiative and tell me,Hey, I completed these multiple things.
That way it's fresh on my mind.
(10:59):
So I didn't get to the see there.
But after I would have completedthat project with the raids, my goal
would've been to work with the teamone-on-one and have them pitch me.
Their successes and then Icould categorize it for them.
Like, okay, what you saidfalls into this or into that.
Do you agree or disagree?
And then teach them how to make theargument for their good works better.
(11:20):
Um, it's delicate to do in business,but it's like a negotiation.
So you actually need topractice it in order to get it.
And this particular company wasopen to that they wanted to hear.
So, uh, that was like theicing on the cake for me.
We didn't get to finish it, butit's something that would've
been proactive for everybody.
The company would've had a very,well, a very articulate staff, which
(11:43):
is needed for problem resolution.
And then with the market as it wasconstantly laying off and whatnot,
this employee would have had theskills sharpened to then go right
into another position if they werelaid off and quickly get another role.
Avery Smith (11:56):
Hmm.
Very cool.
Um, while you're at Tesla, whattools did you use the most?
Lily BL (12:01):
Um, I think I used Jira the most and excel for the validation.
Um, I got heavier into theadministrative side of, of the
software because for Tableau and JiraI was bringing in add-ins to make
them more functional for analytics.
Uh, so for companies, uh, you have toconnect the Tableau software to what,
(12:22):
wherever your data is in the company.
When you use Tableau as anindividual user, you just
connect it to your worksheet.
You can't connect it to somethingelse if you have it, but typically
you just use a worksheet.
So that was different.
And it was a full hostof security clearances.
Um, so I did a little bit of theadministrator stuff, but, uh, JIRA
and Excel round my validations.
Avery Smith (12:43):
That's awesome.
I think that's true.
And, and you mentioned JQL.
Is that kind of like SQL orhow, how are those related?
Lily BL (12:49):
Yeah, so it's very similar to the commands in sql, so that's why I
was able to learn it pretty quickly.
Uh, but then, um, somethings are specific.
It uses a lot of, uh, a lotmore keywords than you would
expect, and they're different.
Um, the software itselfdoes try to help you.
Like it lets you click on buttonsand produces the code for you.
(13:10):
Uh, to an extent, but then youhave to have modifications.
So I would allow the software to allowme to click to build some of the stuff,
but then I would review it and determine,oh, it still needs this functionality,
or this, or this other group of people.
And you would have to manually put thatinto the existing code to make it function
Avery Smith (13:28):
super neat.
So it's basically SQL for Jira, andthey try to make it a little bit easier
for you to actually write the code.
Um, okay.
I'm actually not sure theanswer to this question.
How did you get this job at Tesla?
I remember you messaging me whenyou got the job offer and you're
like, Hey, these are the details.
What do you think?
Should I take this job or not?
You know, it's one of the things I tryto do with my accelerator students, but
(13:49):
I don't remember off the top of my head.
This was a couple years ago now.
Um, how you ended up landing this job?
Lily BL (13:54):
Yeah, I think it was through networking.
Um, at the time I was an instructorfor co-op and I had a cohort that
I would teach in the evenings.
One of my cohort students actuallygot hired by them about a month or
so before I helped them finalize hiswork that they wanted him to see.
So then about a month or so later, I got,uh, contracted by a recruiter on LinkedIn.
(14:16):
I checked with him.
It ended up being the sameperson that contracted him.
So I think what happened is that Ipopped up for her in association to him.
But she never said that.
But that's, that's what wethink the connection was.
And so then she interviewed me.
I actually was like number three orfour, because three or four other
people had said yes and then backed out.
(14:39):
And so then it was really easyto consider, oh, you know what?
I just won't take the job.
You know, it just seems really hard.
But I just kept saying, well,if the, if the manager wants to
interview me, I'll be available.
If the position comes back open.
So after that is how everybodyelse, 'cause there was a lot
of things that it required.
And then I ended up notdoing most of those things.
(14:59):
Uh, so I ended up, uh,hanging in for the interview.
And then in the, in the interview,um, he asked me some questions that
I think everybody else struggledwith and I answered very confidently,
uh, because of the work that Ihad done inside of your bootcamp.
Avery Smith (15:13):
That's, that's awesome to hear.
So, uh, what I was kind of hearing was.
Basically you were, you were connectedto the, to a right person, someone
kind of similar, one of your peers,um, looking to inundate a job.
Uh, and then you had a goodLinkedIn because the other thing
is, uh, on, on LinkedIn, right?
Like you don't get recommended ifyou have kind of a crappy LinkedIn.
So making sure your LinkedIn wasup to date with all the right
(15:34):
keywords, all those projects youhad done inside the accelerator,
I'm sure that helped, uh, as well.
Yeah.
You, you nailed the interview.
Okay, that makes sense.
So what advice would you give to someone.
Who's listening right now who's like, wow,I wanna be cool like Lilly and work for a
cool company like Tesla in the data space.
Like what advice would you give them?
Lily BL (15:52):
Um, I would recommend that they kind of determine what part of the data
portions they like to do, and then afterthey figure out, oh, I like building
the data structures, or the pipelineor the visualizations, dive into that.
I do get, uh, a lot of requests forlike, how can I kick off figuring
out my data stuff and actuallyrecommend them to your free content?
(16:13):
'cause I find it really helpful.
I think you do a good job organizing.
Okay.
You gotta go get the data.
Once you get the data,you gotta clean the data.
Well, once you clean the data, you gottafigure out a quick way to deliver it.
And also the visuals, how youbuild your visuals is gonna kind of
determine what you can say, um, asthe Bluff Fund, like right up front.
I know these are things we do to buildthe projects, but those directly translate
(16:36):
into the interview and also into workingwith, uh, people on site in the jobs.
So if you can find a material that helpsyou hone the skills you naturally want or
like inside of your data, uh, career, itwill make it easier for you to get that
and it will make you a natural to post it.
'cause you'll actuallybe excited about it.
(16:56):
Like, oh, I had a hard timelearning this particular function
in Excel, but I nailed it.
Let me show you guys how I did it.
You'll naturally be like, oh, wehad overtime with sql, but I figured
this out and now I'm gonna post it.
And people actually do look at it.
They might not comment, they mightnot like, but recruiters and also
other people, uh, interested in datawill come and look at your projects
(17:18):
because if you had an issue withit, likely someone else did too.
So if you're constantly posting yourprojects and how you solve the problem.
Uh, they will naturally gravitate to you.
And one thing I always stressis to try to frame my projects
into a problem and a solution.
So the purpose of this projectwas to address this specific
problem and here's the solution.
(17:40):
Maybe they won't care for lookingat the problem, but maybe they're
interested in just a solution.
But that's interesting.
They'll go back and look at theproblem and then read all of the work.
Avery Smith (17:48):
Interesting.
So yeah, projects played a bigrole for you, it sounds like, like
you really believe in, in doingprojects and then posting them on
places like LinkedIn to get noticed.
Lily BL (17:57):
Yeah.
And in the interview for Teslaspecifically, uh, I think the
question that sealed the deal forme was that, uh, the bus had asked
me what I would do in inside of sql.
So he asked me just the generalstuff, like, you know, how would
you get something to come up?
What would you call the tables?
And he goes, it was likehis, his secret question.
It was supposed to catch me off guard.
He says, what if thereisn't anything in there?
(18:20):
Like you asked for it and it doesn'tgive, like there's nothing in there.
What seat we're going to do then.
And we, I had done a module, uh,to the bootcamp, uh, that you have.
And I had picked the short dataset instead of the large data set.
And because I picked the short dataset, my results were different.
And in fact, missing.
Mm.
So I distinctly remember sittingthere for like, what, what happened?
(18:42):
Did I do it wrong?
Rewatching the video, redoingthe thing, and trying and trying
until I got very frustrated.
And then I realized, oh, I pickeda different data set than he did.
So our results are probably not the same.
They're likely missing from mine.
So then I manually went in and checked,and that was exactly what happened.
So when this manager from Tesla asked me.
I knew exactly whathappened when that occurs.
(19:03):
And so I was like, youget absolutely nothing.
It's the most frustratingthing in the world.
Uh, but it's good because youdon't have to keep looking.
There's absolutely nothing there.
You, you're just gonna get a, andbecause I was so confident about it,
having sat in the frustration, helaughed and then was like, I, I think
that she will be able to figure outwhatever she doesn't know and what
she does know will benefit us anyways.
(19:26):
And I think that's what sealed the deal.
So as you're working through theprojects and honing your skills.
Think about what you experience.
'cause that's what's gonna makeyou shine in the interviews.
Avery Smith (19:35):
I, I love hearing that.
I love hearing that the experiencesyou had inside the accelerator
program, uh, worked out wellfor you in, in an interview.
And it's interesting because, uh, Iobviously try to design the accelerator
and we're constantly updating it so thatpeople have less and less problems, right?
Like, we wanna try to make it as easyfor people to learn data as possible.
But the silver lining iswhen those problems happen.
(19:56):
It puts you in a real lifescenario 'cause you're gonna have
problems when you get on the job.
And figuring out how to solvethose, figuring out what's
going wrong, uh, is a skill.
It's kind of a hard skill to teach.
But it's a very valuable skill to have.
So, uh, I love hearing that,you know, a lot of data skills.
I'm curious here what order youlearn them in, and if you have any
(20:17):
tips for anyone who is learningthese different data skills?
Because there's a lot, right?
There's Power bi, there's Tableau,there's Excel, there's sql, there's
Python, there's R Like what order didyou learn those in, and what advice would
you give to someone else learning those?
Sure.
Lily BL (20:28):
Uh, I think the order I learned them in was first Excel
and the Microsoft Office Suite.
Uh, I actually was certified throughthem, um, to use Word in Excel.
However, I didn't understand itas much as I would over time.
So then with the Excel basic knowledge,I was able to navigate most data
and then I realized everything'strickling into information systems.
(20:51):
So when I realized that I went back toschool and I got a degree that focused
in information systems and there I wasintroduced to, uh, data visualizations
where we used a variety of tools.
Uh, the one that stood outthe most to me was Tableau.
So from there I joined an apprenticeshipwhere they used that tool.
(21:12):
'cause it just was visually stunning.
The rest of the stuff could getthe things done, including Excel,
but they were kind of grainy.
But with Tableau.
You could just floor somebodyby just the visual alone.
You wouldn't have to say anything.
They'd just be looking at it for a while.
So I was like, I'm reallyinterested in that.
So I did that.
And while I was in that program,we also covered, uh, Python,
(21:34):
uh, more basics and sql.
And, uh, we also did, uh,presentations, um, of the findings.
After I had, uh, those things under mybelt, I discovered your bootcamps and
then went back to square one with Excel.
Was like, okay, this is how you useExcel specifically for data analysis,
not the other stuff I was doing.
(21:55):
So it redefined, like, it reallysharpened what I knew how to do.
And from there, uh, I went back into sql.
A lot of the companies I worked for didn'tuse SQL as intensively as I expected.
So I was more so, uh,using Tableau frequently.
And then Power bi.
Uh, power bi, um, islike a full stop shop.
(22:16):
For analytics because it allowsyou to do the visual component.
But to do that you need tobe able to pull in data.
To pull in the data, you needto understand like the, uh, the
stakeholder request, and then also howto clean the data and it uses Excel.
So, um, the skills were the samein all of the software you just
(22:36):
clicked in a different spot.
So throughout the software per uh.
Phases or processes.
What I was continuouslysharpening was what is the data
process independent of the tool.
So if I had to start all the way over,the way that I would learn these in
is Excel, uh, power bi and then uh,Tableau and last sql, unless it you
(23:02):
are company that you're targeting does,is focused on sql, I would do Excel
and then sql because if you understandwhat you're doing in Excel, like, um.
V lookup, a next lookup, an H lookup.
They're essentially joining data.
So if you know how to join the data inExcel and you can articulate it, then
you can look at any other software.
Here's um, a sql, let me goahead and join data here.
(23:26):
This is how I do thejoins in this software.
Okay, now I have Tableau,how do I do the joins here?
And you are specifically honing yourskill for joining data, which is like
the backbone for, uh, data analytics.
And then that will parlay youinto engineering if you want.
Uh, but I would go Excel first andthen whatever you learn in Excel,
(23:48):
mirror it in whatever softwareyou can get your hands on next.
I did have the cases sometimeswhere I didn't have certain
software, so I've had to wing it.
Um, I did a lot of G docs andthe, all of the Gmail suite
documentation when for some time Icouldn't afford the office software.
So even if you can't getthe most premium thing.
Do what is affordable, but focus onthe skill you're trying to sharpen
(24:12):
and you'll be able to figure it outeven if you've never used it before.
I
Avery Smith (24:15):
think that's a really cool, uh, point there is like, you know, we use
different software at different times,but really a lot of them do similar stuff.
They get data from places.
You clean data with them, you do somesort of aggregations or analysis or
make some charts obviously, like SQLdoesn't really make a lot of charts.
Like a pivot table in Excel is prettymuch just like a group buy in sql.
(24:38):
Um, so there is a lotof, uh, overlap there.
So that makes a lot of sense.
So Lily, when you were tryingto break into data, there's
obviously a lot of data roles.
Um, there's data analysts,there's business analysts, there's
operations research, which iswhat I used to do at ExxonMobil.
Um, and each one of those jobs,uh, is kind of complicated.
(24:59):
They, they're all data analyst roles, but.
They have different domains,they have different industries,
they have different focuses.
They may use different tools, they mighthave different vocab and, and customers.
So one of the things I really love, um,that, uh, you sent me was like this matrix
you made of a couple different, uh, dataanalyst titles and what you'd be doing
(25:22):
slash what tools you'd be using basedoff of how experienced you you were.
So tell me about this matrix you made.
Why did you make it and, youknow, what does it do for you?
Lily BL (25:32):
So I wanted to share this, uh, with you and, uh, potentially
to anybody trying to break into dataor further career in data, because
this is how I was able to do it.
Uh, pretty much when you startat the beginning, you don't
have a bunch of experience.
Um, in my case, I just knew Excel,but not specific to analytics.
So the way that you leverage, uh, thetool I gave you is that you kind of.
(25:55):
Set up your goals by a five year plan.
And the reason why is because by thefifth year of any profession, you're
considered a professional 'causeyou've been in it for five years, you
have enough working hours to do this.
At a professional level,you're not guessing anymore.
You should know, uh,concretely what you're doing.
So, uh, depending on what kindof analytics you wanna do,
(26:16):
the matrix can kind of guideyou to where you would start.
Let's use me for an example.
I started with Excel and Iwanted to be a data analyst.
My first data rules werenot titled Under Data.
So what I did is that I said,Hey boss, I know you want me to
take care of these appointments.
And it was clerical work, but itwas, uh, handling a lot of data.
(26:36):
So I said, Hey, you have anopportunity here, uh, to figure out
why your patients are dwindling.
So I took it upon myself tooffer a project so that I
can gain the skills I needed.
So in that project I recovered about halfa million dollars, uh, of lost payments
because somebody clicked the wrong button.
And there I secured my Excel experience.
(26:57):
I secured, uh, the patients beingable to return the company, gaining
the money, uh, that had originallybeen lost because, uh, I used Excel.
That's what I needed in order to beginto say, Hey, I have six months work.
With Excel, I have ayear's worth with office.
Um, at the time it was very popularto use the Microsoft Office Suite.
Uh, let's say you secure thethe time you need with Excel.
(27:20):
Now you can say, Hey, in Excel.
I've also executed Pivot chartsand VLOOKUPs so I can join data.
I'm ready to go onto the next thing.
Hey, boss.
Uh.
We have a lot of data ina lot of different places.
We already are integrated with Microsoft,so we can use Power BI to pulling
all the data sets into one location.
(27:41):
Uh, can I get some time to be ableto make that happen so that I can get
you some support with your recordingand then you start figuring that out?
You might, when you, when you do this,you don't have necessarily somebody
coaching you, so you need to relyon the bootcamps or the knowledge
you already have that gives you theconfidence that I can execute this.
If you can't execute in thesoftware you're reaching for, don't
(28:03):
nominate yourself to do the project.
In there, you do it 'cause youalready know you have, you know
how to use that software, but thecompany's just not implementing it.
So then you would jump into Power bi.
Maybe not its most advanced things,but just enough to get your feet wet
so that you can figure out, this ishow I use it, this is what I like.
Once you get in there,you can be like, Hey boss.
Uh.
We're in here with the Power bi.
(28:24):
We have these simple reports, but we havea lot of stuff inside of SQL as well.
I was wondering if you can get me access,uh, to request permission to join them
into the Power bi and that way I canaccess more data and goes from there.
Right.
Well, one of the visualizations inPower BI is a table, so you can actually
organize and clean all of your data insideof Power BI and then export that sheet.
(28:49):
Put it into somethingstunning like tablet like.
It's hard because as you're workingon it, it's not inherently clear
what you're doing, but that's howyou use the document I sent you.
You look at the title you want,what software or what knowledge do
I have now and what can I reach forbased on my hidden skills that I
can start to attribute to my career?
(29:09):
And that's why you slowly grow it.
Now, sometimes the companies will say,no, we don't need any work in Power bi.
We just want it done in Excel.
For me that translated to, I need tofind another company because I really
wanna grow more skills, uh, to get tothe next level because I have five years
to make it to that professional status.
And if I hit five years and I don'thave all the things in my tool belt,
(29:31):
I gotta do more than five years.
That's what I used toget into the next thing.
Um, also, if you don't wanna growyour career, like you're happy with
what you're doing, don't volunteerthe projects or the software.
Um, hone on what you, or focus yourskills on honing what you already
know and that will make you sharperand sharper with what you have.
Avery Smith (29:50):
Well, I think that's one of your skills is that you're really good at
figuring out how to link business to data.
Uh, and I think a lot ofbusiness and operations people
kind of struggle with that.
Um, so it's really cool that youwere able to be like, Hey, I see this
business need, uh, here's how analyticscould help us, uh, in this case.
Um, and I think that, you know,you've done that as well with building
(30:13):
dashboards for, for stakeholders thataren't necessarily, uh, data experts.
Um, I guess how do you have like an eyefor where data can help these businesses
and how do you, uh, help these maybenon-technical, non-data folks be excited
and interested and ready to, to helpwith these data analytics projects?
Lily BL (30:36):
Well, that's such a, that's such a good question.
Um, because you kind ofhave to actively listen.
So, uh, it's almost likespeaking another language.
Somebody can say, oh man, like in real,a real life example, a boss that I had
said, oh, I just want this inside ofExcel, and I'll be happy if we could
just get it from where it is into Excelso that I can analyze it, I'll be happy.
(30:59):
So I got it done.
After it was done, they were sohappy that they decided, I want,
I wish everything can go in there.
And I said, what?
What you want, sir?
Is a warehouse of data.
Mm.
So what he said is, I want everythingin there, or I want this in Excel.
But what they're asking foris an accumulation of data.
They're asking for a pipeline.
If you understand the data portion ofthat, you can translate the regular
(31:23):
English into what that looks like in data.
And that's how you can determineI can fix that or I can give you
something to help you hit that goal.
Or you can determine, oh, you knowwhat, that's just outside of my reach.
'cause your Google and you have a bunchof data, I can't handle that much stuff.
Like I need servers, I need abunch of other stuff, but these
portions I can handle for you.
(31:43):
And that's how you determine I cando this versus I can't do that.
I should offer you this 'causeI know I can execute that.
Avery Smith (31:50):
Lily, that is awesome.
I think that is a superpowerthat, that you have.
Thank you so much for giving a glimpseinto what your career was like, telling
us what it was like to work as a dataanalyst at Tesla and give us some
good, uh, advice and feedback for.
Trying to learn these data skills andtrying to maneuver in our data careers.
Is it okay if we put your, uh, LinkedInin the show notes down below and if people
(32:12):
have questions they can reach out to you?
Sure.
Okay.
Awesome.
Lily, thank you so muchfor coming on the podcast.
It's so good to have youand, uh, good to catch up.
Lily BL (32:20):
Thank you.
Likewise.
Popular Podcasts
New Heights with Jason & Travis Kelce
Football’s funniest family duo — Jason Kelce of the Philadelphia Eagles and Travis Kelce of the Kansas City Chiefs — team up to provide next-level access to life in the league as it unfolds. The two brothers and Super Bowl champions drop weekly insights about the weekly slate of games and share their INSIDE perspectives on trending NFL news and sports headlines. They also endlessly rag on each other as brothers do, chat the latest in pop culture and welcome some very popular and well-known friends to chat with them. Check out new episodes every Wednesday. Follow New Heights on the Wondery App, YouTube or wherever you get your podcasts. You can listen to new episodes early and ad-free, and get exclusive content on Wondery+. Join Wondery+ in the Wondery App, Apple Podcasts or Spotify. And join our new membership for a unique fan experience by going to the New Heights YouTube channel now!
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
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!