Embedded - 120: Boll Weevil Eradication

Episode Date: October 1, 2015

Kathleen Sidenblad discusses her career through Silicon Valley, from engineer at Systems Control Inc in 1976 to VP of Engineering today. For more about Kathy, check out this Storehouse interview.   

Welcome to Embedded FM, the show for people who love building gadgets. I'm Alicia White, alongside Christopher White. Today we're going to talk about careers, going from engineering to VP of engineering. We're going to talk about software in the medical arena. To help us with both of those, I'm happy to welcome Kathy Sidenblatt to the show. Hi, Kathy. Thanks for joining us today. Thanks. I'm glad to be here. Could you tell us a bit about yourself?

I am a very long-time software engineer who has ended up in management, much to my amazement. But I still think of myself as a computer programmer, person who loves to build things. Not gadgets necessarily, but useful applications that didn't exist before I thought them up. That is pretty cool. But it has been a little bit of a while. Do you mind giving times, years, when you started in development in engineering sure yeah no i started programming in the mid 70s i guess when i got out of college and uh have worked at assorted companies as a software engineer mostly the silicon valley kind of companies that were building hardware devices that needed software to help them run. So I worked on software that drove voicemail systems at Octel.

Before that, I was at MeasureX doing process control systems for paper manufacturing. And prior to that, I was at Systems Control, and we did a lot of different think tank stuff. It's an SRI sort of spinoff. I did a lot of different think tank stuff. It's an SRI sort of spinoff. I did a lot of electric power system work there, maintenance scheduling algorithms for electric power plants and modeled the agricultural industry to help decide whether we should eradicate the boll weevil as a country. Anyway, so lots of different interesting software projects. Boll weevil eradication. I thought I've worked on some weird things with that one.

Yeah. Well, I modeled the fertilizer and corn industries, which would be affected if you got rid of the boll weevil. You didn't need to plant as many acres of corn, and so then what would you plant? Anyway, lots of simulation software. So you got a BS in math from Santa Clara? I did, back in the days when there really weren't computer science degrees in many places, especially liberal arts schools like Santa Clara. And then you went to Stanford. I did.

I was in a PhD program in operations research and left before my PhD, left with a master's. Sort of came to the realization that I would probably end up teaching in Kansas or something. They were definitely, the PhD program was around building teachers more than people to work. So anyway, I left and joined industry just to see what was out out there and never went back that's a very familiar calculation yeah i noticed i i gave up on the phd when i noticed that everybody i knew with the same phd was a software engineer yeah well and all the good jobs were taken you know the the the colleges you'd want to go to, but the, you know, podunk you out in the middle of some frozen plane had openings. So, what was your first job? Well, my very first job was, as all good women in the 70s, I was a secretary and put myself

through college because my college, my high school counselor made sure I could type because she wanted me to be employed in my later life, which was useful. So anyway, that put me through school. But my very first job was with systems control after I got my master's at Stanford. And then Measure X and then Octel. Measure X and Octel. That gets us up to the mid-90s. So that was a 20-year span. And those were all very hands-on engineering jobs, right?

Absolutely. Absolutely. But then you went over to the dark side. Tell us about your journey into management. I went into healthcare. And in 1996, I joined a little startup as employee five. We were doing disease management, working with people who are chronically ill. And I was writing software that would page them to remind them to take their medications and building software.

With real actual pagers, not like tweeting at them like you do now. Absolutely. So, yeah, keeping track of when they were due for appointments and things. And we basically had a lot of nurses that called the patients and telephonically helped them with their chronic disease. Because most people who are diagnosed with chronic disease don't get much education or information about the disease. They get the five-minute appointment with the doctor who drops some bombshell on them, and then they really don't know how to manage

their diet and other things to get the best outcome. So anyway, it was a very interesting company. We were sitting right alongside the nurses and building software. And I think everybody else who was there as an engineer at the time stepped back when they needed somebody to manage something. I don't really remember volunteering to manage. It certainly was not a particular goal of mine. I liked being a hands-on engineer. I thought I was a pretty darn good programmer. I really liked building things.

But I came to the interesting realization that if you have a group of other equally talented people and you can get them organized a little bit, it's like having a whole lot of hands and you can be a much better, more effective person if you're driving a lot of work than just a single work. I did always make sure I tried to keep my hands in. And I still write software today, although probably shouldn't as much, but it's probably not necessarily an effective use of my time, but I like to build stuff. Yeah. It was hard at ShotSpotter when I was enough of a manager that I didn't get to code. Yeah, no, that's... Well, and eventually you become a bad manager if you lose touch with the technology.

I mean, and that's really the hard thing. I mean, when you've been doing this for close to 40 years like I have, you really need to always be learning the next new thing and have some hands-on experience. It's not enough to read a book about Ruby and then manage a bunch of Ruby programmers. You kind of have to have some hands-on experience so that you know where the gotchas are. You don't have to know everything that your best programmer knows,

but you need to be able to give them good advice or they won't respect you. So, anyway. And to speak the language. Yeah, yeah. So I've definitely learned a lot of computer languages and, you know, my time. What's your favorite? Database.

Oh, probably C++ is the one I use the most. But I like SQL for power. I do a lot of backend stuff now. I was surprised how much I liked SQL. Yeah. I did not expect to really enjoy databases that much. No, they're very powerful. And doing the right job in the right place with the right technology is really, really important.

A lot of people don't hire any DBAs and hire a ton of Java programmers, and they do everything line by line by line, when you can just slam a whole bunch of stuff through with a sets in the database and you know run in a hundredth of time or a millionth of time so so i was spectating on an argument between c and c plus plus and amused at the people who were adamantly opposed to c C++ because it runs so much slower and embedded systems and everywhere. What would you have said? I laughed. I didn't really do anything. I didn't have a stake in that fight.

Well, it depends on what you're doing with the language, too. I mean, you don't have to use all of the features that are there. My husband is an assembly language programmer, and we started out in very overlapping fields. He actually does a assembly language programmer, and we started out in very overlapping fields. He actually does driver software and stuff and is still programming, again, some number of years. And I've migrated up to the application layer and done a lot of user interface things. So our worlds have gotten increasingly farther and farther apart as time goes on. But he just would never consider doing anything in C++ for the longest time because it was so slow. And he's now using it.

Is it really slower? Virtual functions, okay, I get there's another layer. But it's just one layer. If you go from C to JavaScript, that's huge. But C to C++ is... No, you can do things that are bad in it you know with multiple layers of virtual functions or something but you don't do that but you can use templates and lots of other cool things that are that are there that are

constructs so and definitely wrapping up things as objects has a lot of power somebody out there is screaming at their car radio going no no, no, functional programming, that's where it's at. My husband and I went out to dinner probably in the mid-90s, and we were at a pizza restaurant, and there was an argument going on about the best object-oriented language to use. And I don't remember whether it was small talk and C++ or anyway, there was some objective C, there was some war going on. Some table got into a heated discussion and everybody in the entire restaurant, including the busboys, had a comment about the

languages. And, you know, it's just such a typical Silicon Valley sort of story that you can't talk in the supermarket about your code without getting advice from somebody over the frozen peas. I swear every time I go out to eat, the person at the next table works for Cisco or wants to work for Cisco. There's always a networking conversation. And it's not networking like they say,

you should network with people. No, no, no. This is protocol. Right. So learning languages to keep up to date is important. But going from engineer to manager, did you do any of the win-win negotiation or manipulation for idiots sort of classes? I know they don't usually title it that, but I might go if they did. No, I know. And I probably should have done something, but I never was offered anything. I just was kind of given people and told to figure it out.

And so being a parent was a really good preparation for that. You have the child that doesn't want to do their homework or whatever, and you have strategies for that. You have the child that doesn't want to do their homework or whatever, and you have strategies for that. But yeah, no, and I'm terrible at the politics part of it and the corporate schmoozing to get a good budget and some of those things. I would have probably, probably should have had classes on how to pad your budget so that you get what you need or whatever. And schedules. And schedules. Well, schedules I know. The engineers know how to pad your budget so that you know you get what you need or whatever schedules and schedules well schedules i know the engineers know how to do that so how how do you do schedules

this actually was a listener question recently was how to how to plan a schedule oh yeah well and i mean i i always ask my team what they think it's going to take because it's really really unfair to have somebody say, oh, and this guy will implement this whole massive feature in a week and a half. And no, it's really six months. I don't always believe everything they tell me. And sometimes I ask them why they think it's going to take so long. And we talk about design, but yeah, you sort of start there and you build up the list of tasks. It's hard to get not too detailed, but just enough to have a swag. But, you know, I've been doing this long enough now I have at least some sort of sense. So, speaking of project planning and management, you've seen many of the methodologies and processes that have been proposed and have become popular. What do you think?

Yeah. You know, there's no magic bullet. There just really isn't. You just kind of have to have a reasonable idea of what you want to do. You have to have a reasonable idea of how to do it. And you have to have people that have some hope of knowing how to do it. I interviewed with one company that I didn't take a job at because they had a bug assignment methodology that was like, take a number. So whoever came free next just took the next bug, whether they had ever seen that system

in the software or not. And I mean, their code was just this massive, I mean, they were showing me stuff and stuff was commented out. I'm leaving this cause I have no idea what it does. Kinds of things. I've never done that. No. So,

yeah. But somebody had the good idea of all programmers are just cogs in the machine and they're all equally talented and they all know everything equally well. And that seems to be a very common mistake yeah no and i i have my database guy work on the database and i have my user interface person work on the user interface and it just seems to work better that way well and some cross training is nice for when people go on vacation but you don't have to do it every day. No, it doesn't have to be random assignments. Do you use Agile now?

We do. And, you know, it's probably modified Agile a little bit. I don't make people stand up at 10 a.m. and say things, but we move pretty fast. And honestly, in 96, we were doing very Agile development, sitting next to the nurses saying, what do you need? Try this out. And, you know, we were doing very agile development, sitting next to the nurses saying, what do you need? Try this out.

And doing very tight time circles, getting code out in a couple of weeks and doing many, many releases. And that works better. I have been on the projects of Doom that have the three-year time schedule. That's the ultimate rewrite of some system software. That's always the project that gets canceled because three years they're not going to need that great big thing. You have to do little tiny chunks that make sense and get them out. That's part of Agile I like.

Steering, yeah. So now you are at Amplify Health. I am. And you are VP of Engineering. I am indeed. What does that mean? So, you know, the titles are kind of weird. And in startups, I'm not sure how much sense they make.

In a big company, there's often a system architect and a CTO and a VP of Engineering. And the VP of Engineering is much more worried about the schedules and getting the software written and being accountable for the deliverables. And a lot of the internally facing nuts and bolts where the CTO is a little bit more, maybe the guy who goes to conferences and talks about the cool thing they've built or whatever. And the architect's worried about the design. In a small company, I do all of that, right? So the kind of titles are a little bit meaningless. But anyway, whoever assigned me mine picked VP of Engineering, which was fine.

What does Amplify Health do? Amplify Health works with primary care physicians, and we integrate a lot of data. So we try and turn healthcare organizations into data-driven organizations. And we take a lot of electronic health record data, claims data, device data, and other things, and look for places where people are falling through the cracks and bring that to the primary care physician's attention. Sort of similar to what we were doing at my previous company, Life Masters, but that was a bolt on.

It wasn't the doctor doing it. It was assumed the doctor didn't have time. So they had this other group of nurses who weren't really well integrated with the patient and the doctor. And they would get all this great information and work with the patients, which was great. But sometimes they and the doctors were working at cross purposes because they really weren't on the same team. So it became clear to me we had great technology, but we were injecting it into the healthcare system in the wrong place. So when I got the opportunity to inject it at the primary

care level, right with the people who are working with the patient, and make them aware of things that are maybe going on with the patient that they're not aware of, it seemed like a real win. Well, I knew when I go to the doctor now, like half of the time is spent with him in front of the computer, updating my records and asking if I've had my flu shot. Reminder everybody, get your flu shot. And just, you know, he looks at the computer for a minute and he says, oh yeah, we need to check your levels for that or this or something that's more specific to me. Or he asks about injuries I've had in the past to make sure they're gone. Is it that sort of thing, or is it more specific?

No, it's really a lot more specific. We work often with employer groups, so they will have a whole set of patients, some of whom may not be seeing their physician at all. They might have a chronic condition that they've just stopped taking the medication for and stopped being treated for, and they're in danger of ending up in the ER. Nobody's watching them. Healthcare is really well set up right now to treat you when you show up in the ER with your broken leg. They put a cast on it and they fix you and you go off and you're better. But nobody's watching the people that maybe aren't actively involved in the system or are too actively involved in the system.

Maybe they're shopping around for pain medication at seven different places. And we see all the data from all the sources, sort of the whole patient view. Yeah. So the doctor sees what happens in his electronic medical record, but he doesn't know that that same patient has gone to see four other doctors in the area. And so we can alert people to things that are going on. And, you know, part of it is risking patients and deciding who's at high risk for some sort of an event because they've stopped taking their medication for some mental condition they have or something like that. And a case manager can intervene and perhaps get the person back on track before they crash.

So this is all behind the HIPAA wall. Yes. You have access to all the data. Oh, and there's, I mean, I do a huge amount of paperwork. That's my life um to get access to that and it's all you know got to be highly secure and need to know you know top hush so this seems like it's very personal you can point out to doctors people with individual problems on the other side of the fence there's all these wearables that everybody has now that are tracking all kinds of things and we'll be tracking more things in millions of people.

It seems like the same sort of analysis could be applied to that anonymously to say, okay, this region has this set of interesting data that differs from this region. Why is that? Do you see that coming? Oh, I sure hope so. And I mean, I know companies are working on pieces of that. I really think when the combination of the genomic data, the wearable data, and all of the what are you actually seeing clinically is going to happen, I think that's going to revolutionize where healthcare is going. A lot of times they know a drug works with some people and not with other people, but they don't know why. Is it their diet? If you had their lucky rewards card data as well as their wearable data as well as their genomic data, you could really start to sub-segment the populations.

In the same way you know who's going to buy a new set of towels at Williams-Sonoma or whatever. They've done it in many other industries, and I think healthcare is really lagging behind. And the HIPAA stuff is a problem, and people are rightly worried about the creepy factor. It's bad enough you know I buy towels at Williams-Sonoma, but do you really want a lot of people to know about certain health conditions? And if you get stuff wrong in healthcare, it matters a lot more than somebody gets four coupons from Williams-Sonoma because you can kill people. So it's an interesting business where the data has to be very secure.

You have to be very sure of what you're telling people and not make mistakes. How do you deal with that level of security? Well, we spend a lot of time on making sure our infrastructure is very secure and encrypting things and transmitting things securely. Yeah, that's a lot of my job is to make sure that's done right. Does that decrease the use because it's harder for the physicians to use the data because they have to deal with all of the security? Oh, no, I think it's not that.

Once you get it down to an authorized user, they're seeing just the data that they should see. And then they're seeing it on an HTTPS webpage or something like that. So, it's not onerous for them, and in fact, it has to not be, or it will never get used. Yeah, that's what I was thinking, is that if it was too hard. No, it's tricky. Right.

I had a conversation with a doctor whose relationship to me shall remain anonymous yesterday, where I did hear a lot about... He was on the show. They all know. I was hearing a lot about electronic medical records and how it wasn't quite great yet. No, it's not great. If there are 1,600 of anything that don't interoperate, they haven't solved the problem yet. They have to train on different ones every couple of months and everything changes. So he was talking about buying basically an electronic pen and things just to write things down so he could deal with that stuff later instead of well-being with the patient.

So I think there's... No, the usability is terrible on most of those systems. And yeah, they do get in the way of the patients. And the fact that they don't interoperate is just it's almost criminal and it's it's bad because the amazing things you can do kind of depend on that working well and doctors being engaged with it oh absolutely no i mean think of how the banking system would work if you had to find the ATM for your branch in your state and log in there, and they didn't share data. They had no way of doing it. And that's essentially what's happening in medicine is everybody's –

I mean, banks want to be secure too, right? You don't want to give away people's bank balance to people who don't need to know. But you can send certain messages back and forth between systems that don't violate that and still get certain information out. So they've built so many silos and walls, and everybody is protecting the silos and walls because it's just to the advantage of the vendors who have the big software packages. It's their marketing strategy is to keep everybody else out, which works well. For them.

For them. Yeah. But there's also a huge liability for data loss in medical areas. Oh, absolutely. And so the siloing is half get off my lawn because it's mine, and half I'm building a wall because if anybody gets on my lawn, I'll be in a lot of trouble. Right. Now, there's a law against having a unique patient identifier across systems, which is nuts, right? So all the stuff is set up for billing.

So I know that William somebody got a claim for this, and Bill somebody got a claim for that. Are they the same guy? I don't know. They kind of live in the same city, and they sort of have the same age, but their birthday's one day off in this system from that system. Should I merge their data or not? And if I do, will I kill somebody because I've got the wrong guy's data? And I don't do anything that's going to kill people, but a system that's got blood type information or something, if you're sending it to a hospital, could have some pretty bad.

Yes, please don't screw up my blood type. Right. And that's why they'll always give you a test anyway, even if they think they know it. Is this why medical software is so expensive? You know, I don't know. It's like any enterprise software, I think. The big Oracle databases or whatever, they always have the year and a half implementation time to get things up and the huge price tag. And that's kind of what the huge fact, free EMRs for doctors, web-based. But they're paying for it some other way, right? They're getting advertising or they're collecting your data.

It's never really free. But the fact they don't interoperate is annoying. Sorry, I'm just thinking, before you diagnose this patient, please watch this 30 second ad yeah you and your patient will enjoy this next talk to your doctor about uh wait you are your doctor you went into healthcare related software in 1996 right how has the industry changed over that time? Well, it was not a booming place to be in 96. Certainly, I think some people were trying to come up with companies that would improve the way billing information and things flowed. But once the government started paying doctors to implement EHRs, there was a gold rush.

And all sorts of technical people who said, it's just data. How hard can it be? It's just a UI and had no clue about medicine, which honestly, I didn't when I started either, but I had some very knowledgeable nurses sitting right next to me, beating it into me when I got it wrong. So, you know, I think there was a real gold rush at that point. And that's, I mean, there really were 1600 licensed EHRs at one point. I think there was a real gold rush at that point. I mean, there really were 1,600 licensed EHRs at one point. I think some of them are starting to fall by the wayside because the market really can't support that much diversity. 96 was the year that HIPAA came into action.

Was it? Okay. I just opened the Wikipedia page, so I'm pretty sure about that. Yeah, I know when we first started, it was not in place, and I know we started worrying about it not too long after that. But yeah, and then we had everything with internal servers. We had our own IT room. We had our own walls and everything. We're now on Amazon, right, on the cloud.

And we've made that secure and have signed all the business associate agreements and everything with them that are required. So things have changed immensely. And I think the cost of setting things up and the ability to scale, if we get a new big customer. We used to get in 96 data sets that were too big for any of the disk space that we had available. We just couldn't process it. Now you just spin up

another server in the cloud and there you go. It's amazing to me how much is now in Amazon's data centers. It's a little terrifying, actually. Yeah. I hope they're doing a good job. Yes.

They appear to be. So, going back to your career, because medical stuff is interesting, but what was the longest time you spent at a single company? I spent ten and a half years at LifeMasters. I went from being employee five to

we had 1,200, 1,300 people there when I left. We had nurses in five call centers all over the US talking to patients, running my software, which was very exciting because I'd always been on the project of doom, the two-year rewrite of the system, and spent a long period of time having projects canceled out from under me. But this was really cool, and it ended up not necessarily being a totally happy story for everybody. We didn't make a million or anything, but I think we did some good work, and I think we very much changed the way medical care was delivered. I think chronic disease management

is just a standard of care now, and people didn't even understand what it was when we started. We basically took our Measure X process control for papermaking, saying, you know, if you measure things, the quality gets better, to medical care. It said people with congestive heart failure should weigh themselves every day because if you gain two pounds in a day, you're starting to retain fluid and your heart's going to fail and you're going to end up in the coronary care unit for 10 days. And if we give you a diuretic instead in a timely fashion, you can just stay home and work in your garden and be fine. And so it's just huge, huge cost savings for very little care.

But nobody ever told people that they should weigh themselves every day. And we had them put their weight in the computer and we alerted a nurse if it started to go up. And they'd get a call and lots of things worked better. So it was a novel, and people could not understand what we were doing. I mean, we had the hardest time selling that product in 96. With that story, you've totally sold me the product. Well, it made a lot of economic sense. Yes, because ER visits are expensive,

and a call from a nurse saying, please take this medicine today is not expensive, not comparatively. No. And while hospitals like to make money from surgeries and whatnot, I believe that for the most part, people would rather save people at that level than... Well, it was really a win-win-win. The patient was healthier and happier.

Their disease was better in control. The insurance company saved money. The doctor felt like things were in control and he had some extra sets of hands to help out. And we were collecting data he couldn't get any other way. You go in once a year and you have your blood pressure taken. What's your blood pressure the rest of the time? Who knows? Now there's wearables and other things right that report that um we totally could use that sort of stuff and we did integrate with a lot of scales and glucose meters and other things at life masters so what made you stay so long i mean it sounds like team you were doing fantastic things that would make it worth going to work every day. I had a wonderful, wonderful team. Yeah.

Yeah. And, you know, we were doing really, really cool things. But you get to a certain point in a startup where you're not the right person anymore. We got to a size- From five to 1,200? Where they needed somebody who was politically savvy. Well, I know from my personal experience, you feel like your contribution just shrinks and gets diluted to when you sit there and go, well, they don't really need me. No, when you're in a 10 or 20 person company, it's very clear every day what it is you do. Yeah.

And the contribution you bring. And if you don't come in today, it matters. And particularly when you're doing new stuff and you're able to be thinking up things that didn't exist before, it's very rewarding. Where in a big company, it starts to just become an operational thing. Well, it's another quarter and we need to update the database. And it became more of an IT job which is not really me

that can't possibly be as much fun not as fun as coding and that was where you went from engineer to manager to director yeah I was a director there and then took a vice president's job after that in a smaller company

what was the smaller company? I went to Inclara Health which was doing sort of similar things for end-of-life management. That's a tough area to talk about. It is. So we worked with just excellent, and actually there the nurses reported to me we were such a small company. I had the programmers and the nurses, which is not something necessarily

that you learn to manage. I bet those are different personalities. Absolutely different personalities. And I had to be aware that the nurse just had five of her patients die that week, and maybe she needed somebody to talk to. And probably an engineer isn't the right person to talk to. Here, let me fix that for you. Can we have a meeting and you can talk to your peers and you can have a grief counseling group that I don't have to go to? So, yeah. Anyway, it was an interesting thing.

I learned a lot about controlled substance. We had a pharmacy arm that sold medications that shipped medications to hospices and stuff. So, learned about shipping controlled substances across state lines. Wow, I don't even think it's possible to do that legally so that you obey all of the laws for everybody. Well, everybody's got different laws, that's for sure. So, yeah. We had some really great pharmacists that were working for us there that were on top of the regulations. But yeah, we needed some cool software to keep track of stuff there too. If I were someone else, I would probably ask you for advice on that, but I'm not that person, so we'll just go on. Well, I was amazed to find out that there are grandchildren who steal their grandmother's pain medication to sell it on the black market, which was rather, I mean, not surprising, I guess, but who does that?

Yeah. From there, you were a consultant for a few years. I was. I got kind of burned out on healthcare a little bit. And so I went back and did some consulting work for other companies. And it kind of freed me up at a time when my parents were ill. Did some flexible scheduling things, which was very nice. Did you enjoy consulting?

I mean, that's what we do all the time. So I really enjoyed it. Yeah. I love working on a lot of different projects. The part I don't like is having to go out and sell myself all the time, you know, find the next project. Were you mostly hands-on engineering, programming, coding? Yeah, and some of it was like system architecture review

for big companies that wanted to say, here's what we've got. What don't we have enough of? Or what do we have too much of? So it was a combination of things. But I did a lot of database design for some startups and did some healthcare consulting for some people that just needed advice around HIPAA or other things. And you must not have been driven off from healthcare entirely

because now you're back. No, well, when I got a chance to work in a place that I thought could be effective, where I knew I could build tools that were useful and get them in the hands of the right people, that was very attractive to me. So that was why I chose to kind of come back and do that full time. So going from engineer to more senior engineer to manager, what is the difference? As a manager, it's easier to say, well, this title does this and this title does that. But how do you define the different roles and how do you help your people progress through them?

I think you have to respect the craft. And that's something that has to be really, really important. It's easy to come out of college as a newly minted programmer and you know the syntax of the language. But there's just a lot of stuff you don't know about the best design for certain use cases or the best way to approach certain scenarios. And I think that's what the more senior titles get you. And I really respect companies where they have a career path that lets people stay in a pure engineering role. My husband has chosen to do that. And he's an architect at NVIDIA and does really, really cool stuff. But there's a career path there where they respect

that. And the more senior he got, he gets jobs like talking to the lawyers about the patent portfolio and things that are more senior. You definitely get to go to more boring meetings sometimes. But I think you can be a really cool up-to-the-minute programmer and do something a much harder way than it needs to be done. As you get better at your craft and more and more senior, you've got that experience around it. And that's where I think some of the startups now, where they only hire 22-year-olds, are really missing the boat.

Because I think a mix of people. I think you really need some people who know all the latest stuff, and you need some other people that have just been around the block a few times. Well, and sometimes the latest stuff is just the latest stuff. Right. And it's not actually applicable, it's just, you know, it's what everybody's talking about. It's like any other fad,

and it takes a long time for people to discover whether or not they're actually useful. Well, and things kind of come and go, right? There's always a bottleneck in your system somewhere. And so whether it's memory or network speed or whatever, I've lived through punch cards where everything was run at a central server farm, right? Because that was all you had. Somebody had an IBM mainframe somewhere that you timeshared on. Then you get to the client server

thing and everything's done on the client and then the software gets really bloated. Then mobile comes along and everything's got really tiny memory footprint again and you need to know how to squeeze things back down. Hey, I know how to do that. We used to not have much memory in the old computers. And then everything's in the cloud

and people say, oh, we invented the cloud. It's all, well, yeah, it's maybe not quite as new as, I mean, the clothes are different, but the concepts are the same. So, yeah, knowing how to work around the different restrictions, because you never have enough of something, whether it's processing time, whether it's memory, whether it's disk space, whether it's network speed, you know. Well, and sometimes it's processing time, whether it's memory, whether it's disk space, whether it's network speed. Well, and sometimes it's calendar time.

Right. And so you spend it all in order to get just another day, or you hire another person so that you get just a little more time done. Yeah. So anyway, yeah, you need to design systems that work as best as they can in the current environment. And they'll look totally silly 10 or 15 years from now. Well, why wouldn't that person just use a solid-state disk? Because they didn't have them.

It does require diversity. When I talk about diversity and how important it is to have more diversity in engineering, it's that. Part of it is knowing the diversity of skills, but if you take that a little further, you end up with diversity of viewpoints. Oh, absolutely. And that helps you build a much better system. Engineers building medical, I mean, sure, okay, we read specs really well and we think about things pretty well. But then you put a nurse, you sit down a nurse next to an engineer and it all becomes a lot different. Right. Oh, my favorite story from Life Masters is I had done a beautiful engineering graph of blood pressure with error bars and a very engineering looking

graph. And this nurse came up and said, in nursing school they teach you that in every hospital we graph blood pressure this way with an upside-down triangle and a right-side-up triangle for the high and lows. And pulse goes here and they all look the same and anybody can read that graph. And that was like, oh, well maybe that's the graph I should be drawing then.

Because I was trying to say, oh, no, but look at how cool this is. It's very sexy looking as a UI, but it wasn't the way they just could read it in a glance. And different industries have different languages. Absolutely. Visual as well as spoken it's not always jargon and jargon's not always bad no and life experience is really important too i mean what i know having had my mother pass away and spending lots of time in hospitals and it's a different view of the medical system than somebody who's 20 and you know maybe fell off

their bike once and went to the er and you ER and had a stitch in their head or something. Yeah, well, and you said that kids are a good way to train for management. I hadn't heard that, but it makes sense. So you've been friends with Elizabeth Brenner for a while. For a long, long time, yes. She was a past guest. She was the one I spoke to about the Are You Okay widget. And we were talking about neighbors and making sure that they would pat a stuffed animal and all of that.

She's a good friend and I asked her what questions I should ask you. Because it's always, you know, nice to have your friends tattle on you. One of the things she sent was that you had an interesting interview at OCTEL when you interviewed there. Something about the C++ language? So, I had a couple of interesting interviews there because they had apparently swapped around the schedules without telling people. And so, one of the interviews I had was supposed to be with HR. And so, it turned out that the guy who was actually interviewing me didn't introduce himself. I thought he was HR, because that's what it said on my little time

sheet. And so, he kept asking kind of pointed questions about architecture, and I kept trying to not get too technical on him. So that was an interesting little thing. But yeah, then there was another very young person who asked a bunch of questions about C++, and actually they had the information wrong. So anyway, that was, it's always interesting to figure out exactly how you sidestep that without telling somebody you don't know what you're talking about. So. When, at the last time. We could type it in and see if it would compile. The last time I interviewed for a full-time job, which was a while ago, one of the things I did in the second round was I asked to talk to the person I'd be working with mostly.

And then I asked a number of questions about technical opinions, like C versus C++, that kind of thing. And I kept taking the opposite view. Even though I didn't really care, I just wanted to know that I could argue with him. Interesting. And it took him a little while to recover because he didn't realize. I mean, I said at the end, look, I don't. I'm playing with you. I'm playing with you.

I just wanted to make sure that you weren't the person that stomps off and gets angry. But yes, I also had another interview where they did not know the answer to the question they were asking, and it went really badly. I showed off for the other interviewer in the room and the junior person, wow. Yeah. Don't make the interviewers cry, they don't hire you. No, and don't try and talk to an HR person, to the senior architect. I think he thought, maybe I didn't have quite a grasp of all the details. Well, yeah, because you've interviewed the person who can talk a good general game, but then you push down, you realize they have no idea what they're doing. Right.

He probably thought you were just the person who only does the big stuff well and it is really interesting now because um people pad not pad exactly but fill out their resumes so that it gets through all the automatic resume scanning software so the stuff that comes across my desk has stuff and when i ask people about it things that they say they have done they say i don't know that clearly uh Clearly a little bit padded. And then there are all those interview questions online that people study up on, and they don't actually know the answer, but they've kind of memorized them. So you'll get through a couple of interesting questions and then realize they have no clue. They're totally out to sea. So it is hard to find good people.

Yes. Do you have any interview? I do like interview questions. Do you have any interview questions you often ask? You know, I'm a lot less worried about specifics than whether the lights are on upstairs and people can think about things, you know, in general. So I tend not to ask specific, I know some people have coding tests that they give them or, give them or rotate buffer type questions or whatever, but I tend to try and see if they could figure their way out of the paper bag. You don't ask them the go question, do you?

No. Okay, good. No. More I ask them to talk about work they've done and why it's important. And many, many people have done projects and they have no idea what the big project was, how it contributed to the big project, or why it was important. And I mean, I don't always ding them on that, but it's just nice to kind of know that they're not remotely curious do you suggest they go watch real genius and understand how laser fits into the world and popcorn and yeah anyway and i'm always amazed it big systems you know that that people just have they've worked on their little little piece and

i back up the database. Why? I don't know. You ever restored it? People often put projects on their resume, and then they write them out as if they were instrumental. Right. And so it goes back to what you're saying. They write that they had involvement in this big thing,

and then their involvement was very small. And that's the part of the interviews I actually enjoyed the most, was digging into what people had done because you can very quickly tell if they were being self-aggrandizing. Right, right. Versus, oh, yeah, this person really did do all of these things and understood stuff. That makes me feel a lot better when I get those answers. Yeah, I created a self-driving car. Well, wow, that's cool. Yeah, right.

What part did you do? Well, I screwed the wheels up. Yeah. Okay. That's good. I was instrumental in creating it yeah i think i did reasonably well at a job interview when we talked about toys and for some reason there were toys on the desk and so i took those toys and described how i had built my toys while i played with their toys it was really you yeah, she probably really did do something with toys.

Given she can't keep her hands off it. Yeah. So, Elizabeth also asked, how do you plan your career? How did you plan your career? Yeah, and that's kind of an interesting question. I don't know that I, being a child of the 70s, ever thought about having a career. Definitely, I didn't know anybody growing up whose mother worked. So I know kids now come out and they have a game plan and they have a marketing plan for themselves to build their brand. And I've worked with guys in these San Francisco startups, and they're all very driven,

and they're going to be at this point in their career here, and they're going to job hop four times to get to there. You just want to laugh. And I've never found life works out that cleanly, even if you do have a plan. But definitely, I just wanted to do interesting work with nice people, and that's kind of been my mantra. So I've taken jobs and that was actually advice i got from a senior engineer in my first job she said oh

i just if you can work with nice people and do interesting work that's as good as it gets yeah that's probably good i think my whole plan consisted of get job and that would be yeah i got a first job okay well after that it was yeah was random. Right, so yeah, but people seem very, you know, I'm going to be CEO by the time I'm 35 and I'm going to be on, okay. So it was definitely not my, I guess I wasn't driven, you know, but I've certainly enjoyed everything I've done and have done a lot of very cool and interesting things. What did you want to be when you grew up, when you were small?

When I was small, I read a lot of Nancy Drew, and I thought it would be really cool to be an FBI agent. I didn't know what they did, but I thought they got out a lot more than most of the people in my neighborhood and did more interesting things. So that was my... Do you ever think about getting a PI license? You know, I still read a lot of murder mysteries. So my husband is a little concerned

about the Bludgeoned Husband series that occasionally shows up, but he's safe. Christopher's often concerned by the ads he gets on his Kindle based on the books that I read. Yeah. Yeah. Yeah. Yeah, well, this week they've been all management. I don't know what's going on there.

Oh, no, there was a sale. I regret those books. But there's probably some big data thing with the FBI that, yeah, they probably need me and they just don't know it. I was surprised ShotSpotter, which the gunshot location system, and working on the databases there were just fascinating. Because there was just so much information. And it wasn't about listening to the sensors or anything, because there wasn't that much information from them. But just the sensor locations over time, because they had GPS for precise timing, we could look at those and see the ones in Oakland

and the ones in Southern California and the ones in San Francisco and see the plates shift. Oh, wow. Because you have the data. Right. And the databases can tell you weird things. Yeah. Yeah, that was never a paper we were going to get to write.

No. And it was so sad. But it was so cool. That is very cool. Let's see. What did you want to be when you grew up after college? Did you know then? I went into college thinking I would be a high school teacher

um math teacher that was sort of you know i knew what teachers did so that was probably my my career goal uh but when i ended up in the master's program i just looked for a really interesting job that could use the or skills i had. SCI was a great company to work for there. Lots of super smart people. How big were they? They were probably 250 people. I think everybody had a PhD except like six of us. There'd be another thing I'd add to that.

One or two other women programmers, but almost all of the women were secretaries. It was an interesting place to work. But a lot of people went on from that company to be CEOs of various and sundry places. Just a huge amount of talent and you know a lot of confidence that whatever it is we can do it and that was a great attitude to be infused with at a young age yeah so a lot of people think they have to go out and get a sample code off the web because somebody else has already done it right well you can invent stuff that nobody's ever done before, and that's kind of cool.

Yeah, yeah. But yeah, we wrote library systems and systems to do algorithms to get the most lumber out of logs and traffic light things and boll weevil eradication. I mean, if there was a government grant, we applied for it. It was fascinating. It's nice to see that breadth. Yeah.

And I mean, when I talk about getting to work on children's toys and gunshot location systems, it's like, it's all embedded systems. It's all, there's signal processing, there's small microcontrollers and optimizations. And yet, I get to have all this breadth into all these other industries that's so fun. No, and that's the best part of the kinds of jobs we have, is to get to learn stuff and learn about all different industries. There is life outside the web and apps.

Yeah. Should I ask you about being the only woman in the room or on a team? Or do you have things you want to say? We don't usually talk much about women in tech on the show, but if you've got specific advice. I don't have specific questions, but there is definitely a difference in style. So I know even as early as Measure X days, I was often the tech lead on a project, and I would be the only woman in a room of 24 guys. And, you know, there's definitely a difference in the way people state their opinions or talk over you or whatever. But...

Has it changed that much over... I think there are fewer women and I think the atmosphere is probably less friendly than it was. That is true? I think so. I mean, I worked with Elizabeth and a bunch of other nice women at Measure X, and I think there were more. I spend a lot of time in San Francisco now because that's where Amplify is located, and

we're in startup lofts there, and so I've seen a lot of different companies. But it tends to be the beer pong night you know um guys that that's one of the interview questions is how good are you at beer pong i think um in some of those companies and so if you don't show you if you don't fit in and i certainly wouldn't yeah wouldn't fit in so imagining myself just closing my notebook and politely walking out. Well, anybody over 30, right? Right. Or anybody who has a life, right, if you have kids. But there's definitely a very, and it's a young, energetic vibe.

It's not bad, but it's a little bit short-sighted, I think, sometimes. Yeah. I mean, I do, people say, are there women in tech? And I have found enclaves, essentially groups. It's hard to be the first woman. It's hard to be the second. And then after you're the fourth, it gets a lot easier to hire women. Right, yeah.

No, and I think the diversity is important and the difference in just life experience and what people bring to projects. But, yeah, I don't think it's awful. I don't think women should be discouraged from tech or not even try because they think it's so horrible. I think there's plenty of opportunities. Because we have to work on really cool stuff. Oh, yeah. And we're just as good at building things as anybody else.

Yeah, I think so. So you've mentioned your kids, your daughters? I have two daughters who are in their mid-30s. And how has you and your husband being engineers affected them? That question was phrased differently earlier. That's a good question. You know, you always grow up, I think, thinking that you're a normal person. And I married another person who's very much like me. We're both very analytical and we both like algebra and, you know, the things we think are fun are the things that probably most people in Silicon Valley would think are fun. We do puzzles and stuff like that. But when you see yourself through your children's eyes, you often get a view that perhaps you're not as normal as you thought you were because they'll call you out on stuff.

I have two daughters. One of them is a very talented artist, my oldest one. And my youngest one likes algebra and became a microbiologist. But they're very different kids. And we kept trying to convince the older one that blended coffee bean rate problems are really interesting. Let me explain to you why. Why wouldn't you want to do that? And she would just roll her eyes. They went to a lot of parties with us where

Elizabeth and other equally analytical people were there, and they would just quietly watch. And I know one of my daughters explained to one of her little friends one day that it's okay, I'm just being raised by Mr. and Mrs. Dilbert, And whatever my parents said should just, you know, just let it go. It's okay. And, you know, when you overhear your kid say something like that, you go, huh, aren't all the parents like this? And apparently, no, they're not. So, yeah, I think they're definitely affected. I know my younger daughter cracked up her high school math teacher to where he could not teach for 10 minutes because he was explaining that the plural of vertex was vertices. And she said, oh, yes, in our house we have cleanesses. And just, you know, that's exactly what he did.

He could not speak for 10 minutes. And she said, he thought that was really funny. And I said, why are you telling him this? Because it was kind of an internal home thing, you know. But they probably are. No longer. No longer, no. But you do get called on the fact that, yeah, maybe you're not as normal as you thought you were.

But they've turned out okay. And they have affection for us, I think, in our oddity. Wow, so it isn't entirely genetic. No, I don't think so. And apparently not entirely nurturing. No. No. How do you stay current?

You said you... You're having a dove kit avalanche over there. Yeah, it was due to happen at some point. Okay, let me try that again. You said earlier that you learn new languages and that that's a part of management and keeping current but how else do you stay current i mean blogs magazines yeah i read a lot what i've changed what i read has changed i used to subscribe to a lot of a lot of magazines and i do that less I read various websites and blogs.

I follow conferences about development and stuff just to look at trends and see what's coming. Specifics, please. Well, I mean, lately I've been going to the Strata Conference that O'Reilly throws on big data, because that was interesting, looking at the tools like R and some of the other new data processing things, trying to figure out when do you really need Cassandra or when would you want to use Hadoop or all of those sorts of things. So just trying to stay very aware.

There's kind of an explosion in languages now. Everybody and their brothers got some sort of an open source variant of something. So it's almost getting harder and harder and harder to find what you should be using. When do you use Node.js or whatever? But yeah, I try and keep current. I can't obviously dabble in all of them, but I try and get some hands-on experience with some of the important ones. And definitely I always have tried to build into my budgets some training money. So I tried to make sure that everybody on my team at LifeMasters

could go to a conference or a training class or some kind of enrichment activity every other year. And it was easier to do if they were local, which a lot of things are local here. But we had a little bit of travel money to send people to the DBA conference for SQL Server or whatever. And people just come back energized with so many good ideas, so much broader perspective about what everybody else is doing with the tools. It's well worth it. Which I guess is my cue to remind everybody that the Hackaday folks are having a conference in November,

the 14th and 15th. It's a weekend. It's going to be very low cost, something on the order of bring a hack and you can get in free or otherwise you have to pay five bucks. But it's going to be a whole weekend of technical information. That's great. And if you want to know what technical information, well, the way to know is to propose something,

which I should be doing soon. And you should be doing soon too, because it's kind of cool. All right. That's enough of an ad for them. Christopher, do you have any last questions what do you want to be when you grow up now you know i think i don't want to grow up i just i think growing up you kind of get stayed and and stuck in your ways and i just want to keep keep learning and doing and having a good time and hopefully building things as long as I can. Correct answer.

All right, Kathy, any last thoughts you'd like to leave us with? No, thank you for inviting me. This is definitely one of those outside of my comfort zone activities that's probably good to do every now and again. It is for us, too. Every single week. You've been a lovely guest. Thank you for coming.

My guest has been Kathy Sidenblad, VP of Engineering at Amplify Health. Don't forget about that San Francisco Hackaday Conference in November. Do propose a talk today. Thank you for listening, and thank you to Christopher for producing and co-hosting. If you'd like to say hello or whatever, do hit the contact link on Embedded FM

or email us show at Embedded FM. I will talk to you next week. And until then, I do have a final thought. This one is from Eleanor Roosevelt. The purpose of life is to live it, to taste experience to the utmost, to reach out eagerly and without fear for newer and richer experience.