CTO Wisdom with Cat Miller | Beyond the Program

Hey, listeners, Tim Winkler here, your host of The Pair Program.

2

We've got exciting news introducing our

latest partner series Beyond the Program.

3

In these special episodes, we're

passing the mic to some of our savvy

4

former guests who are returning as

guest hosts, get ready for unfiltered

5

conversations, exclusive insights,

and unexpected twist as our alumni

6

pair up with their chosen guest.

7

Each guest host is a trailblazing

expert in a unique technical field.

8

Think data, product management,

and engineering, all with a keen

9

focus on startups and career growth.

10

Look out for these bonus episodes

dropping every other week,

11

bridging the gaps between our

traditional pair program episodes.

12

So buckle up and get ready to

venture beyond the program.

13

Enjoy.

14

Cat Miller: Welcome

15

Eric Brooke: to CTO Wisdom.

16

My name is Eric Brooke.

17

This series will talk with people

who've led technology at organizations.

18

We'll seek to understand some of the

journeys of a person, explore what's

19

successful, a current problem they're

discovering or digging into, And what

20

are they seeing in the wider tech market?

21

Finally, we'll talk about

some recommendations where

22

they collect intelligence for

them and their organization.

23

I want to welcome to Cat Miller,

um, who I'll be chatting with today.

24

Hey Cat, could you give

me an intro to yourself?

25

Cat Miller: Yeah, absolutely.

26

Uh, thanks, Eric.

27

So my name is Cat Miller.

28

I am CTO of Flatiron Health.

29

We are a health tech company that

seeks to improve and extend lives

30

by learning from the experience

of every person with cancer.

31

32

33

So we're now a wholly owned

subsidiary of Roche Group.

34

Eric Brooke: Could I ask how many

engineers you have at the moment?

35

about 400 are tech, um, or tech in

36

some fashion, uh, probably 250 of those

are traditional software engineering

37

and the rest are, uh, security, I.

38

T.

39

Data science, um, those

kinds of flavors of roles.

40

Thanks, Kat.

41

Eric Brooke: Um, so let's talk

a little bit about your journey

42

to your executive career.

43

Tell

44

Cat Miller: us.

45

So, um, I guess how I got into technology

in the first place was a bit by accident.

46

So I feel like I've been falling

into roles my whole life, um,

47

you know, as a math person.

48

And so I went to school for math,

um, and, uh, decided that math wasn't

49

fun and ended up in technology.

50

Uh, for a long time, I

didn't want to be a coder.

51

I was like, ah, it seems awful to

have to code all day, which is ironic

52

because now it seems like it would be

actually really fun to code all day.

53

Um, and then I, I was

in a number of startups.

54

So my, my path has been startups

and healthcare startups, uh,

55

56

And we were a growing company.

57

I was quite a few years

on the road at that point.

58

And so I was, you know, voluntold

into management as one does

59

when you're a growing company.

60

And had the privilege, honestly, to,

um, grow in that space as the company

61

got bigger, eventually, um, becoming

VP of one of our business lines.

62

And, uh, a year and a half ago

becoming CTO of the company.

63

Cool.

64

Thank you for that, Kat.

65

Eric Brooke: What would you say is the

difference between, say, non exec roles,

66

say the director role versus an exec role,

say the VP or the CTO roles that you've

67

Cat Miller: both done?

68

Well, it's interesting.

69

I actually thought that, um,

the CTO role would just be

70

like the VP role, but bigger.

71

Um, as a VP, I probably had 120 people

under me, and I felt like, you know,

72

what's the difference, 100, 400?

73

Um, and I actually feel like those are...

74

Really different roles.

75

Um, I think that, uh, things that I care

about a lot more now than I did then.

76

I think a lot about budgets.

77

Um, it's not sexy, but it's real.

78

It kind of the buck stops with me.

79

I think a lot about, like, aligning

different parts of the company.

80

Certainly that's part of the role

at different levels, but I think

81

this, the extent to which that is a

primary part of my job has shifted.

82

Some of it's about just being a good team

member to the rest of the exact team.

83

So being a thought partner and

working with them on projects.

84

And then I think there's this element of

like external, um, both visibility, but,

85

but also actually more importantly, like

understanding the external landscape,

86

which is again, something that you can

and probably should do as a director or

87

VP, but is now kind of, it feels like a

much more substantial part of my job to

88

make sure we're not missing out on what

we should be doing from the ecosystem.

89

Awesome.

90

Thank

91

Eric Brooke: you.

92

Um, In terms of that journey, what

would you say helped you in that

93

journey, kind of transition from

the non exec to the exec roles?

94

Cat Miller: Um, I definitely got

some feedback along the way, uh, that

95

frankly I didn't, I didn't like, and

I didn't adjust well to at first.

96

Around, um, my approach to problem

solving, I think, like many, I was

97

used to being, if not the smartest

person in the room, someone who,

98

like, had a really strong opinion.

99

And so I would come into conversations

with a pretty fully formed view of

100

what I thought the outcome of that

conversation should be, and it didn't,

101

I was right, a decent percentage of

the time and what I needed to learn

102

is that didn't, that didn't matter.

103

Um, it, what mattered was that I

wasn't making space for other people

104

and I wasn't making enough room.

105

To, uh, bring others into the

fold and have those conversations,

106

you know, authentically and in

a way that made them feel heard.

107

Um, and so I think that was a huge

adjustment that I had to make in order

108

to show up as an executive is to kind of.

109

Basically be able to

roll with things better.

110

Um, I think another flip side of that is

also, uh, I think we, we at Flattern, we,

111

we have this, uh, value, which is think

yes before no, which I hated when we first

112

rolled it out, which I think is really

exemplifying of my problem, which is that

113

I, again, I thought I had, like, I thought

I knew what I needed to do and was like

114

very set on like, this is my book of work.

115

This is my team's book of work.

116

And part of what I had to.

117

Adjust to is like, actually, my

job is to accomplish what my boss

118

needs from me, and I have a lot

of autonomy and how I do that.

119

But when they tell me, actually, we really

need to do this thing because it's really

120

important for, you know, patient privacy,

or we need to do this thing because, um,

121

that's what our customers really need.

122

I need to be willing to, like,

roll with it and say, okay,

123

what does this mean for me?

124

And how do I.

125

Reprioritize and that's something

I think, um, was a challenge

126

for me in the beginning.

127

Thank you.

128

Yeah, I

129

Eric Brooke: definitely

traveled some of that journey.

130

Um, I, I remember that phrase, be

curious, not judgmental is a really, even

131

though you have that thought process,

um, I often found if I'm really good at

132

something and I stay really good and keep

doing it, that leads doesn't leave space

133

for other leaders to fill that space.

134

So, yeah, I think, uh, a good

aspect of that journey is really

135

understanding those concepts.

136

Cat Miller: I think it's also really

relevant, um, when you do skip levels or

137

have conversations with other people in

the organization who sometimes, you know,

138

start, start with the big guns, right?

139

This is horrible.

140

Or I really, you know, this

thing is really problematic.

141

Uh, it's, it helps to

depersonalize a little bit to

142

bring that curiosity and say.

143

Okay, great.

144

That is your perspective.

145

And now I'm really interested in getting

to the bottom of why you feel that way.

146

Um, I think it's like, just, I mean,

curiosity is such a good tool, but, um,

147

it's, it's important to learn to wield it.

148

Well, yeah,

149

Eric Brooke: and it's a bit like the

power of your words as an exec becomes

150

so much more powerful than you realize.

151

Um, to people and also getting

feedback from people is harder as an

152

exec from say, the average engineer,

not all engineers, but like some.

153

Cat Miller: Yeah, it's funny.

154

I was actually just saying to someone

yesterday that I feel like I'm really

155

happy with where I am with that right now,

which is that because I have the CTO role,

156

when I say something, people like listen

to it and like have to pay attention.

157

But I feel that I'm very lucky that

my team is also in a place where they

158

don't feel like they have to jump to it.

159

So it's not like cat said, um,

oh, we should be, you know,

160

expanding our snowflake usage.

161

So everything has to be snowflake.

162

They, they, like, respect me enough

to process it, but they respect

163

themselves enough to say, you

know, what does that mean for me?

164

Or maybe I shouldn't do that.

165

Or maybe I'm not the right person.

166

So I feel very, um, very

blessed, honestly, at the

167

moment to be in that situation.

168

I've definitely observed.

169

Situations where the exec is the, you

know, 10, 000 pound gorilla and comes

170

in and says something and, you know,

wreaks havoc on all of the roadmaps.

171

So, uh, it's a delicate balance that

I'm, I'm always like very careful

172

to, um, to sort of test the waters.

173

That's

174

Eric Brooke: great.

175

Cause it feels like you've built

psychological safety and have people

176

that also have opinions of their own to

kind of give you that back and forth.

177

Awesome.

178

Um, let's, um, step into what got

you into technology to begin with?

179

Like, why did you become

a software engineer?

180

Talked a little bit about

math, but like, why software?

181

Cat Miller: Um, so I,

like I said, I liked math.

182

And when I was in high school, um,

that meant that, you know, I was

183

certainly looking at tech schools

for, um, for where I go to school.

184

Uh, and MIT just felt like a very

safe option, um, in the sense of I

185

was never going to regret going there.

186

Um, and so I went there thinking, Oh, I'll

be, you know, maybe a math major, maybe

187

I'll do a little philosophy on the side.

188

Um, and then I got there and I realized

that, um, university level math is

189

very hard, um, and it's very different

from competitive math and from, you

190

know, the kind of practical stuff

you learn in, you know, say calculus

191

or Dickey Q or something like that.

192

Uh, but what was true at the time.

193

So I was in school in the

early two thousands is that

194

33 percent of the school.

195

Was in the computer science

department, so 30 percent 33 percent

196

of students were had at least 1 of

their majors as computer science,

197

including, like, all of my friends.

198

And so it was almost like a cultural thing

that I just, like, took the intro class

199

because everyone took the intro class.

200

It was like popular.

201

Um, and I was like, oh,

this is kind of fun.

202

Um, and I would not say I ever

really, there was never like a plan

203

to get in technology into technology.

204

There was just a, well, I

guess I'm taking these classes.

205

I guess I'm majoring in this thing.

206

I guess if I major in this thing,

that's maybe what I'm doing for my job.

207

But I've never been confident if I

looked more than 2 years out that

208

it was necessarily going to be

what I was going to be doing it.

209

Just I also was, I think,

lucky to be in at a time.

210

If you think about, like, the early

211

But once we recovered from that, and

212

up until very recently, there's been

like a constant growth in technology.

213

So it's always been a very,

I'll say, easy place to be.

214

Yeah.

215

Eric Brooke: Um, awesome.

216

What was, um, probably one of the

surprising things that you found in

217

whichever aspect of your career that

you want to talk about, um, that you

218

thought, Oh, I didn't expect that.

219

Cat Miller: Um, I think I

didn't expect, uh, the extent to

220

which the more senior you get.

221

The less where you came from

in terms of function mattered.

222

So in my early career, I was, I

think I mentioned, I wasn't very

223

excited to be coding all the time.

224

Um, more out of ignorance

than actual dislike, I think.

225

And it took me, it took me a while

to realize that being a manager of

226

engineering, particularly being a

director of engineering, yeah, you

227

have some different responsibilities

from being a director of product.

228

But as you get more senior, your, your

roles kind of converge and you are

229

still a leader and you are still, um,

helping to shape something and you

230

are still thinking about strategy.

231

And so I think what, uh, what I didn't

know in the beginning was, I didn't know

232

that, like, you know, starting out in a

coding, you know, starting out as coding

233

would eventually lead to a very similar

place than starting out as an analyst

234

or starting out as a product manager.

235

Um, I'm not saying that my role is exactly

the same as a chief product officer,

236

nor that I could do it as well as a CPO,

but there's a lot of similarities, and

237

I just didn't kind of realize that how

much that role would shift over time.

238

Eric Brooke: Cool.

239

Thank you.

240

So let's talk a little bit about what does

success look like in an executive role

241

and what has helped you be successful?

242

Cat Miller: Yeah, I mean, I think that,

um, success for me, I mean, first of

243

all, something that we've started to

do, I think, much better in my company

244

in recent years is, uh, actually have

really clear, um, success criteria.

245

And so my success criteria right now

are actually not the success criteria

246

would have had last year and probably

not the ones I will have next year.

247

Um, currently, I would say

I'm, there's always the kind of

248

expectation that engineering is.

249

Delivering and, uh, you know,

not slowing down the product.

250

Uh, we're not having a lot of instability.

251

There's sort of, I would almost

call that table stakes engineering

252

also that I'm filling my roles.

253

Um, there's a certain amount of like.

254

Org maintenance that is expected.

255

Interestingly, although if any of

that stuff was going wrong, I would

256

certainly be called on it and rightly so.

257

I think then a lot of ways

what people evaluate me on is.

258

The things beyond that.

259

So how well am I controlling costs

in that context of being able to

260

continue to continue to deliver?

261

Um, and, you know, a project that

I, and I think many CEOs have

262

have embarked on this year is.

263

For example, how are we making

sure that we're not missing

264

the boat on generative AI?

265

So when something new and

transformative comes out, how do

266

we enable the entire business?

267

So the tech org, but also the non

technical parts of the org, um, how do we

268

help them kind of ease into using that?

269

Um, and so those are my success criteria,

you know, this year, you know, next

270

year will maybe be similar, but I

think that those criteria could easily

271

change over time with the shape of the

business and what the business needs.

272

Yeah.

273

Eric Brooke: So I hear you.

274

So you're talking about that, like,

actually your success is based on what

275

does the business need based on the

economy, the businesses in, um, with

276

the progression of technology, which

is consistent, um, that there's always

277

something new coming out and can it help?

278

And generative AI is definitely

one of those leaps, um, which, um,

279

we're all evaluating at the moment.

280

Um, how do you know your

team has been successful?

281

Yeah.

282

Cat Miller: Well, I might say being

successful and healthy are like related

283

but different, so I might talk about.

284

Both, um, you know, my team being

successful to me, this is kind of back

285

to tech doesn't slow down the business

or tech is enabling the business.

286

I have never found metrics that I

find perfect for this to some extent.

287

The metric I like is.

288

Are the managers and do the and

the product managers, especially do

289

they feel like we're moving quickly?

290

Like, if I ask them, how

fast does your tech team.

291

You know, move and I give them,

like, 3 options, like average

292

below average or above average,

like, are they above average?

293

Um, or at least average.

294

So there's a certain amount of, um, I

think very much art to that, um, and

295

sort of a sense of like, how difficult

are the problems and how fast are

296

they kind of moving through them?

297

How fast are they unblocking?

298

Um, when I think about is my team healthy

or and these are related, of course,

299

because I think 1 of the things that

is a challenge and can hurt both of

300

these is when you're, you're blocked

a lot when you have a lot of like.

301

Things that are not serving you as a,

as a business, whether they are like

302

flaky tests or whether they're, you

know, uh, lots of, uh, hurdles to jump

303

through, lots of blockers to work through.

304

So for me, I get a lot of

value from just conversations.

305

Um, for me, this is certainly with my skip

levels, um, and asking them very probing

306

questions, you know, Uh, you ask someone

how are things going and they say, say,

307

okay, but you ask them, what's the worst

thing that's happening to you right now?

308

Or, you know, what's the worst team

your team is facing your team is facing.

309

You get a different answer.

310

Um, so I like using really, uh,

provocative questions that force

311

someone to tell me what the

worst thing that's going on is.

312

I also, I have a chief of

staff, I have an HRBP who, uh.

313

It's nice to have other people who

will sort of funnel me sentiment, um,

314

you know, that other folks might feel

more comfortable to approach them

315

or let them know what's going on.

316

And obviously, they're not going to,

um, you know, remove it on a minute

317

anonymity from those people, but they

will tell me sort of the, you know,

318

what they're hearing on the ground.

319

So I, I, some people talk about walking

the halls, I think, in a hybrid or remote.

320

World, we're really not talking about

that, but I, I do think it's, um,

321

leveraging every conversation I have to

get a sense for how are things going?

322

How are people feeling?

323

Like, what's what's the mood in the room?

324

Eric Brooke: I really like the, the

balance of success and health, um, because

325

obviously that also feeds into retention,

um, that people are enjoying the work

326

they're doing and they feel good about it.

327

They're more likely to stay.

328

Um, would you say, so in terms of

success and health, how would you

329

interrelate performance inside that

330

Cat Miller: as well?

331

I've definitely seen,

uh, success be hampered.

332

By performance, and I mean,

success, obviously, at an individual

333

level, but also at a team level.

334

And so, uh, something that I definitely

think has shifted for us, uh, in the last

335

year or so is a real focus on particularly

leaders, um, managers, but also, I

336

think, especially directors saying you

are accountable for how well your teams

337

are doing and how, how, like, successful

they are, what their, what their pace is.

338

Um, I think what's

interesting about that is.

339

Uh, direct, you know, directors

sort of definitionally have a number

340

of teams under them, and I think.

341

In many cases, if a director were

to sort of, they'd have an ability

342

to stack rank their teams and

say, this one was really fast.

343

This one, like, not so much.

344

And you, you want that to be within

kind of a narrow set of parameters.

345

And if it's not, you

want them to address it.

346

So I think it's really important

to hold directors accountable for.

347

Uh, the basically the pace of

their teams and then let them

348

hold the managers accountable.

349

Um, you know, do the root cause

analysis, figure out what's

350

going on in some circumstances.

351

It really is like, the thing they're

doing is really hard or the technology

352

they're working in is really hard or.

353

We haven't given them, up skilled them

in the things they need to be working in.

354

You know, this team is working

in React for the first time,

355

so of course they're slow.

356

Um, I think that's all the stuff

that you can find out on the

357

ground that, that I can't know

from a, from a 10, 000 foot view.

358

But if I hold my directors accountable

to it and my VPs hold my directors

359

accountable to it, like, that's, I think,

been the sweet spot of, of, of that sort

360

of, like, performance, um, accountability.

361

Eric Brooke: Awesome.

362

Thank you.

363

Um, so is there something that you're

trying to figure out at the moment

364

or something you're kind of delving

into that you're would like to talk

365

Cat Miller: about?

366

Yeah.

367

So I think where we are as a company,

um, we've been around for over a decade.

368

I've been there for over nine years.

369

And like I said, my

background's in startups.

370

And so I've been through this company

as a startup for a lot of years.

371

Uh, We were acquired, but we were

acquired as a wholly owned subsidiary

372

and given sort of the advice,

like, keep doing what you're doing.

373

Like, you're on a good path.

374

Um, we want you to expand your reach and

there's a, there's a set of incentives

375

for us for a startup for someone, uh,

for a company that is, you know, pre

376

IPO or looking to be acquired, which is

really about speed and getting as big a

377

market as possible and not so much, at

least historically on unit economics.

378

Um, and I think, you know, this story may

change given everything that's happened in

379

the last year, but I would say very much.

380

Historically, you see companies.

381

Like Uber and others, like burning

money, but they have, like,

382

really high customer growth.

383

They have really high revenue, and

that's what folks are looking at.

384

So we were in that stage for a long

time, and now we are a mature company

385

that is no longer in that sort of,

we need to fight for the market in

386

order to, for our right to exist.

387

And we need to be thinking much

more about that unit economics

388

that like, what is our business?

389

Like, what does our business end up

being shaped like at the end of the day?

390

Um, and those are different decisions.

391

Um, is very, very rational as a, as a

startup to throw a lot of money at compute

392

and not worry about, um, your hygiene

around, uh, your AWS, you know, bill

393

necessarily, because one, your AWS bill

probably isn't even that high because

394

you don't have that many engineers to

spike it up, but also you need that time.

395

You need the engineering time to do

other things, and you don't need it to,

396

like, make sure you're tearing down a

handful of EC2 instances or whatever.

397

Um, I'm, I'm being like different

people will have different

398

philosophies on this and I'm not

saying there's a right or wrong.

399

I'm saying that is a, a type of thing that

happens and your mileage may vary as to

400

what's happened in any particular startup.

401

But at some point that all comes

home to roost when you realize

402

that you have spent years with.

403

That sort of attitude or that sort of

philosophy, which, again, may have made

404

a lot of sense and now you have a mature

business that needs to look a certain way.

405

And so my challenge, and it's my

company's challenge as well, but I

406

think it's mine, particularly in the

tech space, because tech is a very,

407

you know, it's a big cost center.

408

Um, it's a very expensive part

of running an organization.

409

Is really how do we make that transition

from, uh, having things that are hard

410

to maintain or things that maybe weren't

built with longevity in mind into a place

411

where I mean, 1, there's less frustration.

412

Like, I think I talked about

cost, but I think frustration.

413

Is it is an interesting cost proxy

and that when people are blocked,

414

they are not being productive.

415

Um, so how do we smooth the paths, make

it easier, have more robust systems and

416

technology at a, at a lower cost point

per revenue, like per dollar of revenue.

417

Um, that's like, that's the

challenge that we're working through

418

right now as an organization.

419

And it's, I think a lot on me to figure

out how I nudge the organization.

420

Cause these are also scary things.

421

Like I am not talking about

necessarily like removing people.

422

But it's hard to talk about cost without

getting that, um, you know, getting

423

your, your hackles up a little bit

about, like, are you, are you cutting?

424

Are you making a smaller?

425

Like, what does this mean?

426

Um, and I'm really talking about,

you know, controlled growth, um,

427

which I think is really healthy.

428

Uh, but there's definitely a, a, uh,

a shift that has to happen in the

429

organization to, to get to that place.

430

So

431

Eric Brooke: that's really helpful.

432

So you're talking about like that

stage where you're really scaling and

433

growing to some degree about efficiency.

434

And now you're trying to work

out what does efficiency look

435

like for us as an organization?

436

Um, and that would include, I think,

like, onboarding, how long things take.

437

Is this part of our system too complex,

so nobody can actually change it easily?

438

Um, give us some, um, can

you give us some specific

439

Cat Miller: examples?

440

Yeah, well, actually, it's, it's

interesting you talk about onboarding

441

because something that, that made

me realize when you said that is,

442

we've done this in some places.

443

We are extremely efficient

at onboarding, recruiting.

444

Um, we were in hyper

growth for many years.

445

And so something we are very good at is.

446

Bringing new engineer, you know, hiring

new engineers, identifying them, um,

447

bringing them in the door and getting

them set up, uh, is something that we

448

actually like invested a lot of energy.

449

And because that's actually what we

had to like, that's what we, where we

450

ran into scaling problems in the past.

451

So we have optimized things, um, maybe

something, but, but we haven't necessarily

452

done that in other parts of our system.

453

So, um, you know, maybe an example

is, uh, I think that infrastructure

454

has been, I think, a challenge in

the organization for a long time.

455

And there's a number of reasons for that.

456

I was an early engineer and I can say,

I am not an infrastructure expert and

457

I remember feeling the pain of, well,

I don't really know how to do this and

458

there's not really a best practice.

459

There's not a way that we

do this in the company.

460

So I will just do whatever works.

461

And when you bring up 20

systems that were each done.

462

With someone with varying degrees

of competence, you get a pretty, um,

463

pretty diverse set of systems with,

for example, diverse monitoring.

464

Um, and so, uh, you know, we've, we've

been in the process and are more or

465

less at the end of the process of

having all of our infrastructure be

466

code through things like terraform.

467

But I think what is now becoming

true is, um, we're now moving to to

468

Kubernetes and an EKS solution, which

abstracts that problem even further.

469

So, you know, I say we go from, we'd go

from the phase of, uh, pushing buttons

470

in an AWS console to get things up to,

uh, using something like Terraform, you

471

know, infrastructure as code all the

way to manage infrastructure, central

472

organization that, you know, maintains it.

473

Um, leveraging companies like AWS, who

are better at running infrastructure

474

than we are, um, and honestly getting

like cost savings, both in terms of

475

the machines, but also in terms of

the human maintenance as part of that.

476

So.

477

I don't think that's a unique journey

and other companies will have done

478

it at different points, depending

on their needs, but I think that's

479

an example of, uh, the kind of, uh,

progression that, that we're on.

480

Awesome.

481

It's an

482

Eric Brooke: exciting set of problems.

483

Very different, as you

say, to the earlier stage.

484

We don't often get to, um.

485

Were there any, um, other kind

of areas as you kind of like step

486

stages that, um, so you talked about

infrastructure, you talked about

487

onboarding is really, really good.

488

Cause we got really good at that, um,

other things that you haven't got to that

489

you want to kind of explore a bit more.

490

Cat Miller: Well, I think we need

to continually get better at what it

491

means to have a platform organization.

492

Um, I know that this is hard

because I read a lot of blog

493

posts about how hard it is.

494

Um, and I think this is a place where

we've made a lot of progress, uh, but we

495

as a company have at least 2 and depending

on how you count it, 4 or 5 distinct

496

businesses that are really very different

from each other and built on, uh,

497

different technologies for various reasons

that are not going to change anytime soon.

498

And so.

499

It's been a challenge historically for

us to understand what can profitably be

500

moved centrally and what really needs

to stay close to those businesses.

501

So some things end up being easy.

502

So, uh, you know, we use get lab,

uh, that maintenance being the same

503

across all those organizations.

504

You know, that's a core

piece of technology.

505

Um, that makes sense to centralize

and no one's raising their hand and

506

saying, gee, I really want to own.

507

You know, I want to own that.

508

Um, but other things get more complicated.

509

Even, you know, I just talked about E.

510

K.

511

S.

512

Um, you know, ownership of of E.

513

K.

514

S.

515

Clusters.

516

Uh, if we're gonna have sort of one

central cluster, then, of course,

517

we would manage it centrally.

518

Does it make sense?

519

Are there different S.

520

L.

521

A.

522

S.

523

And different requirements for

different parts of the organization?

524

Um, and it's not a unique example.

525

I think, um, that's one of the really

interesting areas of how do you, how

526

do you enable platform organization

to be very effective and impactful

527

on these different business units

who have very different needs,

528

uh, without the business is this

business unit sort of feeling like.

529

Oh, I really needed a thing

and I couldn't get it.

530

So screw that.

531

I'm going to do it myself.

532

Uh, there's, there's always this

like tension, um, that I think has,

533

has gotten a lot better over time.

534

And I think part of that

is getting ahead of things.

535

So, so now our platform organizations

are at a place where they're

536

building things that are better.

537

They've had enough time to be

able to get ahead of the problems.

538

And so now they're building things

that make the other engineers go,

539

wow, instead of building things that

make the other engineers go, why

540

didn't I have this six months ago?

541

Um, and so that's been, I think,

a major, like, tipping point and

542

transformation of, like, making

things easy and, and, and good rather

543

than just making things possible.

544

Um, so that's been a pretty, like,

exciting evolution and one I think we very

545

much need to continue and think about,

again, what things belong centrally and

546

what things belong, you know, very, very

close to the teams that are using them.

547

Eric Brooke: Awesome.

548

Thanks Kat.

549

So we talked a little bit

about your journey, your

550

team, what you're looking at.

551

So I wanted to kind of step into what

does your interaction with the board

552

look like, um, to kind of get a bit of

flavor, because that's often a mystery

553

to many people before they get to kind of

554

Cat Miller: CTO role.

555

Well, I'm not sure that my

board interaction is typical.

556

Um, we're a wholly owned subsidiary,

which means that our board is composed

557

of Roche, um, executive members.

558

Uh, and honestly, in a lot of cases, that

means that board interactions are best

559

managed by our CEO and COO, um, because

these are folks who are perhaps closer to

560

us than, than in other board situations.

561

Um, and I will also say that being in

the healthcare space and the health tech

562

space, um, technology is not the only, or

even necessarily the most important thing.

563

Um, so our board is really interested in

what problems we can solve, and they are

564

also very representative of our customers.

565

And so a lot of, um, I think a lot of

like more standard board interactions for

566

a more like classic tech company with a

more classic board would look different.

567

Um, but, but often I'm, I'm, you know,

observing or, or, um, just kind of

568

being aware of what's going on and, and

making sure I'm clear on what are the

569

constraints we're operating on, um, you

know, really ingesting that sense of what

570

is the context, um, but letting the CEO

or COO kind of take the lead on, you know,

571

financial reporting and other things.

572

I would say more recently, everyone's

interested in a I so, um, you know,

573

making sure that we're, uh, doing a good

job of talking to not just our board

574

members, but, but, you know, generally

folks about what our strategy is there.

575

I think that has been a

little bit of a shift.

576

Um, and there's been a, like, a little

bit more interest in having tech, um,

577

take a leading role in talking about that.

578

Um, but I think this is so what I,

what I take from that is that this is

579

the thing that will shift depending

on, you know, where the risky parts

580

of the organization are, or where the

focus areas of the organization are.

581

Um, but I think we're very lucky to

have obviously very friendly, um, board

582

who, who is very close to our business.

583

Eric Brooke: Awesome.

584

Thank you.

585

Okay.

586

So moving kind of outside,

um, what are you seeing in

587

the wider kind of tech market?

588

You can.

589

Determine the constraint

of its health tech office.

590

Why do you call?

591

Um,

592

Cat Miller: I mean, in the

wider tech, uh, market.

593

I mean, this is not new information.

594

Definitely.

595

There's a slowdown.

596

And I think this focus on efficiency

that I'm talking about is.

597

Being felt more broadly, I certainly see

it in startups that I'm advising around

598

around their run rate and other things.

599

So, um, I also think that, uh, you know,

engineer salaries have been going up

600

substantially for the last 20 years.

601

And I've always wondered

how that will end.

602

I'm not saying that it's ended, but I

do think that this is put in renewed

603

pressure on, um, what, what do you

need a software engineer to do?

604

Or what do you need?

605

Like, a really expert engineer to do?

606

And what can you do in other ways?

607

So I think the quality of low

code and no code solutions

608

has come along dramatically.

609

Now, it's still not necessarily

what you're going to use to

610

build a production system.

611

But I think, uh, there are places

where in the past you've needed

612

engineers and tech folks as glue

where maybe that's becoming less true.

613

Um, it's more possible.

614

Google has tools to build, you know, apps,

uh, with no code or low code solutions.

615

Um, You know, just within

their own ecosystem.

616

A lot of folks are coming out with that.

617

And even I would say, chat GPT has made

it much easier for someone who has a

618

little bit of background to get all

the way to having something usable.

619

So I definitely think there's going to

be, isn't going to be more of a trend

620

of how do we use engineers for the

things only engineers can do and how

621

do we kind of shift, um, Shift other

kinds of work sort of down to or, you

622

know, sideways to other functions or

other types of humans closer to the

623

closer to the business stakeholders.

624

To be honest, it's a

it's a good direction.

625

Um, so that's, you know, I think

definitely a thing that I see, um, I

626

think health tech is really interesting

and probably we shouldn't even talk

627

about it as one thing because there

are wildly different components of it.

628

I definitely see a dynamic where,

um, there's a, there's a decent

629

number of companies that, uh, sort of

started with a like tech enabled play.

630

Um, so we're gonna make this better,

um, through tech and what they end up

631

building is kind of a concierge service.

632

So I can think of a couple

companies that are basically saying,

633

here's the thing that's hard.

634

Um, you know, in our tech, in our health

ecosystem today, uh, we want to make

635

it easier and like the, like the model

that we have is basically like humans.

636

Humans can help.

637

And then we're going to use like tech

to make those humans more efficient.

638

Um, I've seen more like, that seems

to be the play that's actually

639

winning at the, at the moment

in a lot of places because it is

640

possible, um, you know, because like.

641

You, you can make that thing

better with humans and, and you

642

probably can make those humans

a little bit better with tech.

643

Um, it's not sexy.

644

It's not exciting.

645

Like most people aren't like, great,

I'm going to like, you know, make this

646

call center of nurses, like, you know,

um, able to do a better job, but I,

647

but I do think that's where a lot of

opportunity feels like it is right now.

648

What I, where I hope we get to and

where I think we are on the brink of

649

is, uh, using tech to like fix a lot

of the glue in the infrastructure, so.

650

Um, take an insurance company, like

if you're an insurance company and you

651

have lots and lots of humans pouring

over lots and lots of documents, um,

652

you know, not only does that, you

know, money because you're paying

653

those people, but it takes time

because humans take time to do things.

654

And I think the more we can ease those

kinds of, uh, interactions, or, or maybe

655

another 1 is that is more relevant to flat

iron, um, clinical trials, uh, when people

656

are on clinical trials, their doctors are

manually entering data into a database.

657

Um, and it's the same data in many

cases that was also contained in

658

the, the electronic health record.

659

So they're basically doing double entry.

660

That is obviously inefficient.

661

It makes practices unable

to run as many trials.

662

Um, it makes trials less

appealing, uh, it leads to errors.

663

And so there are all these

opportunities for, um, I think

664

technology to, uh, speed up.

665

Again, kind of unsexy glue things

that are in in the interstitials

666

of of our healthcare ecosystem.

667

Yeah, and I

668

Eric Brooke: think customers would

appreciate a faster healthcare

669

system, a faster insurance system.

670

Absolutely.

671

Um, thanks for that, Kat.

672

So let's talk about what's helped

you grow, um, or helps you grow and

673

scale as you've traveled your journey.

674

Like, what are the things that, um,

you would think were helpful for other

675

CTOs or VP of Engineering or directors

with ambitions to become executive?

676

Cat Miller: Um, it's interesting how

I was thinking about like books that

677

I've read that really resonated for me.

678

And I realized that the book that's been

resonating for me lately is actually a

679

book I read seven or eight years ago,

which is called switch by Chip and Dan

680

Heath and it's how to change things.

681

Um, and I read it, like I said,

quite a while ago, but the thing

682

that sticks with me is this idea

of looking for bright spots.

683

So when you want to drive change,

it's very hard to go from 0 to

684

1 and also eat unless you are

like, deeply emesh in the problem.

685

Your solution may not make any sense.

686

Um, and so what you want to do instead is

look for bright spots in the organization.

687

Um, and that idea I use on a daily

basis at this point, um, and it's

688

funny because when I first read it,

I was like, yeah, that makes sense.

689

But it didn't, um, it didn't

feel like it was part of my DNA.

690

Like, I didn't think this is a

book that I'm going to then be

691

referencing, you know, a decade later.

692

And so, uh, I think first of all,

that specific book I think is great.

693

And then I'm really grateful for a CTO who

for a long time gave all his underlings,

694

um, just like books he was reading, um,

and sort of, you know, in a very subtle,

695

you know, in that sort of, um, social

engineering way of like, if I give you

696

this book, then you have to read it.

697

Um, I'm grateful for sort of being forced

into consuming some of that material

698

that I wouldn't have done done by myself.

699

Um, so that I really appreciate.

700

I would say now a thing I do is

whenever I'm introduced to a CTO who

701

is at a company is similar to my scale.

702

I absolutely pump them for information.

703

I become like 20 questions.

704

I, um, I have like, you know, a whole

host of questions that I will ask and

705

then I will try to get them to be like,

I will try to get an email so that I can

706

come back to them with questions later.

707

And obviously I'm, you know, so

it's a 2 way street, but, um, I.

708

Yeah.

709

I very much value having

peers who are facing similar

710

problems at other organizations.

711

It's helpful to understand

the lay of the land.

712

It's helpful to understand

what isn't isn't working.

713

Um, and it's comforting, you know,

sometimes you're at, you're in this job

714

and you're like, wow, these things aren't

going well, or why is this thing so hard?

715

Everyone else, like, talks about it.

716

Like, it's super easy.

717

And then you talk to CTOs that have

gone through it and you get the

718

real story behind it and not the

sort of varnished public version.

719

And you realize that there is

actually, um, you know, there's,

720

there's like, a lot of nuance there.

721

It can be really helpful.

722

So, um, I was told this many years

ago and I didn't take it seriously.

723

And now I do, you know, build a peer

network, um, make sure that you have.

724

Folks that you relate to, to talk about

any of the challenges that come up and,

725

and that you can email and say, like, how

do you think about geodiff compensation?

726

Um, you know, those are, uh, networks

that provide that are great too.

727

Um, but I, you know, that's something

that that's been really important for me.

728

Eric Brooke: Are there specific

networks or communities in New

729

York that are helpful for future

tech leaders or tech leaders?

730

Cat Miller: Um, 1 community that I'm

a part of that I've gotten a decent

731

amount of value from is 1stmark, um,

has a CTO guild network, uh, which,

732

uh, they're pretty good about finding

people and reaching out for their events.

733

And that kind of gets

you on the mailing list.

734

But, um, that they have basically

a community and a mailing list that

735

has been really good for again, these

kinds of questions of, um, there is a

736

thread this week about snowflake spend,

which is, Very interesting to me.

737

Also, even though I was not the one who

sent it, I find that almost universally

738

the questions and answers there

end up being really relevant to me.

739

Um, you know, what's the ratio of, um,

PMs to engineering in your organization?

740

Um, it's a great way to get, like,

a lot of different answers, um, to,

741

to, like, these kind of fundamental

questions and kind of gut check yourself.

742

Um, so that's been

that's been really nice.

743

Um, and that.

744

You know, I think a lot of the connections

I've made this year have really been

745

about looking at events and making kind of

informed decisions about where do I think

746

I'll be able to, you know, connect to

people who, who, um, are running companies

747

that are sort of similar scale to mine.

748

Um, and might have sort of a similar ethos

or similar interest in, in, um, sort of

749

exchange in that kind of information.

750

Cool.

751

Eric Brooke: Um, one of the question

Kat for you, um, we don't have to put

752

this in, but I'll ask the question

and see if you feel good about it.

753

Cause I didn't add it in the

list is, um, what do you do

754

Cat Miller: for fun?

755

Ooh, what do I do for fun?

756

Um, I'm a competitive archer,

so I'm usually at the archery

757

range a couple days a week.

758

I live in New York City, and I can

tell you that no one should pay to

759

live in this city without taking

advantage of something about it.

760

Um, for me, the cultural

institutions that matter are, I

761

love Broadway and I love theater,

and so I'm also a big theater nerd.

762

I think last year I probably saw

almost everything that was on Broadway.

763

Um, and, uh, so that's like a,

something that, um, has been really

764

important to me to continue to enjoy.

765

Living in this vibrant and bustling,

but also very challenging city at times.

766

Eric Brooke: Kat, I really appreciate

your time today and sharing your

767

journey and sharing what you're seeing.

768

Um, thank you very much.

769

Thank

770

Cat Miller: you.

771

Tim Winkler: Calling all

startup technologists.

772

Have you ever dreamed of hosting your own

podcast, but don't know where to start?

773

Well, here's your chance to shine.

774

We're thrilled to introduce Beyond

the Program, our exclusive miniseries,

775

and we want you to be a part of it.

776

As tech leaders and mentors, you'll

get the exclusive opportunity to

777

become a guest host right here

on The Pair Program podcast.

778

Share your expertise, insights,

and stories with our audience of

779

startup focused technologists.

780

Excited?

781

We knew you would be.

782

To be considered, head over to myhatchpad.

783

com backslash contribute.

784

Fill out a brief form

and submit it our way.

785

Let's co create something

amazing together.

786

Don't miss this chance to elevate your

voice and expand your personal brand.

787

Visit myhatchpad.com/contribute.