Latest / What's Your Problem? / Can Robots Fix Recycling?

Transcript

Can Robots Fix Recycling? · What's Your Problem?

0:15
Speaker 1: Pushkin. Imagine you're running a factory. You've got to put
0:22
Speaker 1: out a consistent, high quality product that your customers will buy,
0:27
Speaker 1: but you have no control over the raw materials that
0:30
Speaker 1: come into your factory every day. One day, half of
0:34
Speaker 1: what comes in is literal garbage. The next mixed in
0:38
Speaker 1: with your usual inputs is some random lithium ion battery.
0:42
Speaker 1: It's a fire hazard and also a surfboard weird. And
0:48
Speaker 1: you have to deal with all this stuff and still
0:50
Speaker 1: keep getting your product out the door. This is literally
0:54
Speaker 1: the way the recycling business works. Recycling plants take in
0:59
Speaker 1: a largely random, occasionally hazardous stream of stuff, a stream
1:04
Speaker 1: of stuff that changes in a pretty unpredictable way from
1:07
Speaker 1: day to day, from hour to hour, and then recycling
1:10
Speaker 1: plants have to turn that random stream of inputs into
1:14
Speaker 1: aluminum and plastic and cardboard that other companies will buy
1:18
Speaker 1: and use to make new stuff. This is why my
1:22
Speaker 1: guest today calls recycling the most demented form of manufacturing
1:27
Speaker 1: on the planet, and it's why she and her colleagues
1:29
Speaker 1: are trying to use technology to bring some order to
1:32
Speaker 1: the recycling chaos. I'm Jacob Goldstein, and this is what's
1:42
Speaker 1: your problem the show where I talk to people who
1:44
Speaker 1: are trying to make technological progress. My guest today is
1:48
Speaker 1: Rebecca who Troms. She's the co founder and CEO of
1:51
Speaker 1: a company called Glacier. Rebecca's problem is this, how do
1:55
Speaker 1: you use AI and robotics to make recycling a somewhat
1:59
Speaker 1: less demented business. If Rebecca and her colleagues are successful,
2:03
Speaker 1: they'll not only help recycling plants work better, they'll help
2:06
Speaker 1: companies figure out how to recycle more of this that
2:09
Speaker 1: they're sending out into the world in the first place.
2:12
Speaker 1: Our conversation started with Rebecca talking about the moment around
2:15
Speaker 1: five years ago when she and her co founder decided
2:18
Speaker 1: to start the company.
2:21
Speaker 2: I was, you know, even at the time, obsessed with trash,
2:25
Speaker 2: just obsessed with where does all of our stuff go?
2:28
Speaker 2: And it's one of those I call it like a
2:29
Speaker 2: Matrix moment or a red pill moment where once you
2:32
Speaker 2: realize that you've never thought about where all of your
2:35
Speaker 2: garbage goes after you put your bins on the curb,
2:38
Speaker 2: you can't unsee that, right. And so this was also
2:42
Speaker 2: around the time where there was a lot of change
2:44
Speaker 2: happening in the recycling industry. So we're rewinding to roughly
2:47
Speaker 2: twenty eighteen twenty nineteen. One cataclysmic shift for the industry
2:51
Speaker 2: is that China, who had previously been the world's largest
2:55
Speaker 2: buyer of recycled feedstock to make into new things. They
2:59
Speaker 2: basically said very rapidly, you know what we're taking in
3:03
Speaker 2: the world's recycling, but most of this is trash. People
3:06
Speaker 2: are not sorting it well enough. We're getting a ton
3:07
Speaker 2: of contamination, and we don't want to end up as
3:10
Speaker 2: the planet's landfill or incinerator. So we're going to drastically
3:14
Speaker 2: increase the bar on quality of what we're accepting. And
3:18
Speaker 2: then that caused shockwaves throughout the globe and certainly for
3:21
Speaker 2: the US, where suddenly recyclers for the first time in
3:23
Speaker 2: a long time, were like, the game is not to
3:26
Speaker 2: just crank through all this recycled material, bail it and
3:29
Speaker 2: ship it overseas. We actually need to invest a lot
3:32
Speaker 2: more in increasing the bar on quality and on purity rate.
3:38
Speaker 2: And what's even more challenging is that a lot of
3:40
Speaker 2: the backbone, historically and even to this day, is still
3:43
Speaker 2: just people standing next to conveyor belts sifting through our
3:46
Speaker 2: recycling in our trash and of course that's not only
3:50
Speaker 2: very dangerous, it's not a very well compensated job. There's
3:53
Speaker 2: a lot of hazards there. But also there's this massive
3:55
Speaker 2: sort of labor shortage.
3:57
Speaker 1: Yes, it seems like a robot friendly moment, a robot
4:00
Speaker 1: friendly environment. So like, what's your move, what's your first move?
4:03
Speaker 2: So my co founder had already at the time, he
4:06
Speaker 2: wasn't my co founder, he was just a friend of
4:08
Speaker 2: a friend. He was already pretty intent on this idea
4:11
Speaker 2: that hey, you know, automation AI, all of these technologies
4:15
Speaker 2: are so good now that for the first time, we
4:18
Speaker 2: could feasibly rapidly commercialize a purpose built industrial robot, specifically
4:24
Speaker 2: for recycling sortation.
4:26
Speaker 1: Huh.
4:26
Speaker 2: And it's not going to cost us massive amounts of capital,
4:29
Speaker 2: and we can actually do it in a matter of
4:31
Speaker 2: like a couple of years.
4:33
Speaker 1: So this idea that oh, now is the moment? Is
4:35
Speaker 1: it computer vision? Like, what is the underlying technology that
4:39
Speaker 1: made five years ago or whatever? The moment when oh,
4:42
Speaker 1: we can do this in a way that hasn't been
4:43
Speaker 1: done before.
4:45
Speaker 2: Yeah, Honestly, it's the confluence of a lot of things.
4:47
Speaker 2: So I'll break it down into sort of the computer
4:49
Speaker 2: vision piece and then also the hardware or the robotics piece.
4:53
Speaker 2: When you think about where industrial automation has come from
4:57
Speaker 2: even to this day, a lot of those technologies are
5:00
Speaker 2: operating in really well defined, truly repetitive roat environments. So
5:06
Speaker 2: think about a robot at a warehouse, and it's literally
5:08
Speaker 2: just palatizing identical boxes over and over again. And so
5:13
Speaker 2: to harken back to this idea of recycling plants as
5:16
Speaker 2: being this extremely volatile manufacturing environment, even if you have
5:20
Speaker 2: automation that's just sorting let's say aluminum cans, right, You're
5:25
Speaker 2: talking about aluminum cans that your computer vision needs to
5:28
Speaker 2: detect in infinite varieties, not just thinking about the wide
5:33
Speaker 2: variety of cans that are on the market in all
5:35
Speaker 2: of their colors and designs, but also the fact that
5:38
Speaker 2: they show up not as pristine cans but crinkled in
5:41
Speaker 2: various ways, stuffed into bags. Like, there's so much heterogeneity
5:45
Speaker 2: that even just identifying that item on conveyor belt that,
5:50
Speaker 2: by the way, has dozens of other types of items
5:53
Speaker 2: on it is already a massive challenge that only recently
5:57
Speaker 2: has been something that we can adapt to in a
5:59
Speaker 2: cost effective way. And then you can layer onto that
6:02
Speaker 2: the fact that now you're not only seeing those items
6:04
Speaker 2: with this computer vision system, but you also need to
6:08
Speaker 2: find a way to actually go and grab that material
6:10
Speaker 2: and sort it into the right location. So when we
6:13
Speaker 2: talk about advances in sort of off the shelf parts
6:17
Speaker 2: that you used to design and make your robot, or
6:19
Speaker 2: even like the gripping technology that's available, a lot of
6:22
Speaker 2: that even a decade ago, would have required an immense
6:26
Speaker 2: amount of r and D, with a much bigger team
6:28
Speaker 2: and a much higher price tag to get to the
6:30
Speaker 2: same point that we've gotten to after just a couple
6:32
Speaker 2: of years.
6:33
Speaker 1: Tell me about the first one you built. Tell me
6:35
Speaker 1: about building a prototype and putting it in the world.
6:37
Speaker 2: Man, So there are a lot of different stages to
6:40
Speaker 2: our prototyping. The first prototype, if you want to go
6:43
Speaker 2: way back, was when I hadn't even decided to start
6:46
Speaker 2: this company with my co founder yet, but I did
6:48
Speaker 2: tell him I would help him learn about the industry
6:51
Speaker 2: see if there was some sort of a business to
6:52
Speaker 2: be had, and we met in his kitchen in San
6:56
Speaker 2: Francisco where he had a little corner set up. So
6:59
Speaker 2: like very very simple. I think that it actually involved
7:02
Speaker 2: a used yogurt tub as like this rotating wheel with
7:06
Speaker 2: a piece of string tied around it, Like that's how
7:09
Speaker 2: janky we were talking, but it worked and we were like, Okay,
7:12
Speaker 2: there's something here. The first piece of equipment we actually
7:15
Speaker 2: installed into a recycling facility also tells you a lot
7:18
Speaker 2: about the constraints that these facilities are under. I was
7:21
Speaker 2: kind of thinking we had to have this super built out, sophisticated,
7:26
Speaker 2: polished thing that we've proven out to the nines in
7:29
Speaker 2: the lab, and we actually called up a number of
7:32
Speaker 2: recycling facility operators nearby and one of them was like,
7:35
Speaker 2: you know what, when you got something, just like bring
7:37
Speaker 2: it in here and try it out, because literally I
7:40
Speaker 2: remember him saying, if your robot can pick one more
7:43
Speaker 2: can than I would have gotten otherwise, Like it's already
7:46
Speaker 2: worth it to me.
7:47
Speaker 1: Just try it. Was there a rat moment, tell me more,
7:51
Speaker 1: did you see a rat?
7:54
Speaker 2: Literally?
7:54
Speaker 1: I was like, what is that not a metaphor not
7:56
Speaker 1: a metaphor a rodent?
7:59
Speaker 2: I have seen many rats. Literally, Even on that first
8:02
Speaker 2: install there was a moment where I was like literally
8:05
Speaker 2: army crawling under a conveyor belt to fasten one of
8:08
Speaker 2: the legs of the robot, and I came eye to
8:11
Speaker 2: eye with a rat, who then, of course ground a
8:13
Speaker 2: piece of food that was on the floor and then
8:14
Speaker 2: scurried away. Right, there are many friendly critters running around
8:18
Speaker 2: some of these facilities.
8:19
Speaker 1: Was there any part of you that kind of loved it?
8:22
Speaker 2: Oh, one hundred percent.
8:24
Speaker 1: So let's talk about where you are today, both on
8:27
Speaker 1: a kind of micro level, like what your robot or
8:29
Speaker 1: robots look like, and then also a little more macro
8:31
Speaker 1: of like the scope of the business. Do you have
8:34
Speaker 1: like one basic robot what to look like or you
8:37
Speaker 1: got a bunch of them?
8:38
Speaker 2: Yeah, so we have one base model of robot. We
8:42
Speaker 2: actually are already working with several dozen customers across the country.
8:46
Speaker 2: But if you imagine any sort of conveyor belt in
8:49
Speaker 2: an industrial facility, our robot think of it almost like
8:52
Speaker 2: a table, right, So it's got four legs and it
8:54
Speaker 2: kind of sits over that belt, and then the guts
8:57
Speaker 2: of the robot or the mechanisms doing the picking are
9:00
Speaker 2: kind of over the top of that conveyor system. So
9:03
Speaker 2: you've got these arms that are going back and forth.
9:06
Speaker 2: They can pick up something from the belt and then
9:09
Speaker 2: they carry it off to one of those sides of
9:11
Speaker 2: the belt. Where they actually drop it into the right location. Okay,
9:14
Speaker 2: so that's the robot. And now one other thing we
9:17
Speaker 2: haven't really talked about is this computer vision system.
9:19
Speaker 1: So that's yeah.
9:20
Speaker 2: Imagine basically like a camera with some lights to illuminate
9:23
Speaker 2: the belt sitting on this little rig that's a little
9:26
Speaker 2: bit upstream of the robot. So the material passes under
9:29
Speaker 2: the camera, the camera has a second to process or
9:33
Speaker 2: a should I say, a couple milliseconds to process what
9:35
Speaker 2: it's seeing, and then not only does it tell the
9:38
Speaker 2: robot you know, hey, there's a can coming in this
9:40
Speaker 2: location picket and then put it into this other spot,
9:43
Speaker 2: but what we're also finding is that that data as
9:46
Speaker 2: a standalone is also able to massively advance these operator's
9:51
Speaker 2: abilities to understand and optimize their facilities. So that's opened
9:56
Speaker 2: up a whole new world of use cases.
9:58
Speaker 1: Because they weren't gathering data in that kind of way before.
10:01
Speaker 1: They only sort of knew what was coming through in
10:04
Speaker 1: a very gross macro way.
10:07
Speaker 2: Right, And this gets back to you know, recycling facility operators.
10:10
Speaker 2: I have so much admiration for how they have gotten
10:13
Speaker 2: really resourceful with trying to understand their operations. But you know,
10:16
Speaker 2: to give you a sense of things. The state of
10:18
Speaker 2: the art in the industry to this day is still
10:21
Speaker 2: mostly manual audits. And when I say that, I mean
10:23
Speaker 2: imagine taking a half ton of material off your line
10:27
Speaker 2: and then literally having two to four people hands sort
10:30
Speaker 2: and categorize and wigh each item to understand what's coming through,
10:34
Speaker 2: and then assuming that that half ton is representative of
10:37
Speaker 2: like the thousands of tons coming through your facility on
10:39
Speaker 2: a yearly basis. I often tell the story of one
10:42
Speaker 2: of our early data customers, this gentleman who runs a
10:46
Speaker 2: recycling facility in California. When he met me, he was
10:49
Speaker 2: telling me that he had mounted a goat pro camera
10:52
Speaker 2: above his conveyor belt, the one that was basically supposed
10:55
Speaker 2: to be all trash leaving the facility, but he knew
10:57
Speaker 2: he was missing some good stuff, and he would spend
11:00
Speaker 2: an hour a day after work going frame by frame
11:03
Speaker 2: through some random snippet and manually tallying how many cans
11:07
Speaker 2: and bottles were on that line, and then using that
11:10
Speaker 2: to back into what he would try and change in
11:12
Speaker 2: his operation the next day, and then he would check
11:15
Speaker 2: that day to see if it changed anything. And so
11:17
Speaker 2: imagine his delight when I told him, hey, actually we
11:19
Speaker 2: can mount our own camera on there, and suddenly we'll
11:22
Speaker 2: just give you access to a dashboard.
11:24
Speaker 1: A machine will literally count everything that goes.
11:27
Speaker 2: Literally in real time. So we're seeing that, you know,
11:30
Speaker 2: the robot is this incredible foot in the door with
11:32
Speaker 2: a lot of our facility partners, but that everyone's starting
11:36
Speaker 2: to realize that, hey, actually this data can also help
11:39
Speaker 2: us understand the entire world, not just the location where
11:42
Speaker 2: the robot is sitting either.
11:43
Speaker 1: So at this point is it kind of a robot
11:46
Speaker 1: business on the front, but really you're like a computer
11:49
Speaker 1: vision data AI business.
11:52
Speaker 2: So a lot of customers come to us saying, you know,
11:54
Speaker 2: I literally had a gentleman tell me a couple months
11:57
Speaker 2: ago like I would never pay you to tell me
11:59
Speaker 2: what I already know about my trash. And I'm like,
12:01
Speaker 2: I'm not going to convince you that you don't already
12:03
Speaker 2: know everything about your trash. But you know you want
12:06
Speaker 2: a robot, Let's get you a robot. And that has
12:08
Speaker 2: sense of all the conversation where we're just sort of
12:10
Speaker 2: starting to show him this data and he's like, oh,
12:12
Speaker 2: actually I didn't realize that this was the case. The
12:14
Speaker 2: flip side is also true where someone's like, I don't
12:16
Speaker 2: know if I need a robot yet, but I'm really interested
12:18
Speaker 2: to see what I'm losing on the back end. We
12:21
Speaker 2: install that camera and then suddenly, lo and behold that
12:23
Speaker 2: data makes the case that, holy cow, I'm losing so
12:25
Speaker 2: much stuff. I don't just need one robot. I maybe
12:27
Speaker 2: need two or even three robots. Right, So it's kind
12:30
Speaker 2: of this mutually reinforcing flywheel that's been really integral to
12:33
Speaker 2: the success of our business so far.
12:36
Speaker 1: Tell me about your work with Amazon and with Colgate Palmolith.
12:39
Speaker 1: What are you doing for them with them?
12:41
Speaker 2: Yeah, So to start, maybe just I'd love to explore
12:45
Speaker 2: this idea of what is a circular economy because it's
12:48
Speaker 2: a buzzword that gets thrown out a lot, but it's
12:51
Speaker 2: really important to understand why Amazon and Coliate and their
12:54
Speaker 2: peers matter here. Right now, we are living in mostly
12:57
Speaker 2: a linear economy. In other words, someone makes a thing,
13:00
Speaker 2: we consume a thing, and then we dispose of it. Right.
13:02
Speaker 2: A circular economy tries to turn that process into a circle.
13:05
Speaker 2: So instead of throwing it out in a landfill forever
13:08
Speaker 2: or incinerating it, that material gets brought back to the
13:10
Speaker 2: front end and reused somehow to make new stuff that
13:13
Speaker 2: we can then consume. And the ideal is to make
13:15
Speaker 2: this go on forever so that we limit our resource consumption.
13:19
Speaker 2: So now that we have this growing base of recycling
13:22
Speaker 2: facilities that are you know, gathering data, that are getting
13:25
Speaker 2: a better understanding of what's coming into their facilities, what's
13:28
Speaker 2: actually being bailed and sent out to the end markets,
13:32
Speaker 2: We're working with companies like Amazon and Colgate on a
13:35
Speaker 2: number of fronts. You know, the first is even just
13:37
Speaker 2: to understand where is all of that packaging going to
13:41
Speaker 2: their credit They and several of their peers have realized
13:45
Speaker 2: that there's a paradigm shift possible now from we have
13:49
Speaker 2: designed a thing that's technically recyclable, you know, our packaging
13:52
Speaker 2: R and D engineers have made this awesome mono material
13:56
Speaker 2: HDPE toothpaste tube in Colgate's example, to now trying to understand, okay,
14:02
Speaker 2: well we made this thing that is recyclable, is it
14:04
Speaker 2: actually getting recycled? And that was a lens that we
14:08
Speaker 2: couldn't really get at scale before and so now with
14:12
Speaker 2: Glacier's technology, we're able to monitor in real time, you know,
14:15
Speaker 2: how much of these tubes are actually ending up at
14:18
Speaker 2: the recycling facility, and once they're in the facility, are
14:21
Speaker 2: they ending up in the right place, are they being
14:23
Speaker 2: sorted correctly such that they can actually be turned into
14:26
Speaker 2: new stuff, or are they ending up in the landfill,
14:29
Speaker 2: in which case you know, suddenly this recyclable tube isn't
14:32
Speaker 2: very recyclable at all. So we're starting to answer these
14:34
Speaker 2: really really critical questions.
14:37
Speaker 1: So Colgate knows whatever how many tubes of toothpaste they
14:40
Speaker 1: sold in a city, and if you are working in
14:43
Speaker 1: the recycling facility for that city, you can actually count
14:46
Speaker 1: how many tubes of that toothpaste came down the recycling
14:49
Speaker 1: line and how many tubes of that toothpaste wound up
14:51
Speaker 1: in the bin. I mean, is that the reductive version
14:54
Speaker 1: of what you're.
14:54
Speaker 2: Saying, essentially? Yeah, And where we're even seeing now, like
14:58
Speaker 2: with these rapid advances in AI and in detection, that
15:02
Speaker 2: first of all, it's no small feet to even define
15:04
Speaker 2: what is a tube and how do you tell the
15:06
Speaker 2: difference between a toothpaste tube versus a sunscreen versus a
15:09
Speaker 2: lotion tube. But now we're getting to the point where
15:12
Speaker 2: we can actually say the brand of toothpaste, it is
15:16
Speaker 2: like from all of those visual markings, and so we're
15:19
Speaker 2: just seeing this sort of Cambrian explosion of interest from
15:22
Speaker 2: a wide variety of different you know, packaging producers and
15:25
Speaker 2: brands to really start understanding this previous black box on
15:29
Speaker 2: what happens once they release this packaging into the wild
15:32
Speaker 2: for consumers to buy.
15:34
Speaker 1: So, I understand that California has a law that is
15:38
Speaker 1: in some fashion supposed to put companies like on the
15:41
Speaker 1: hook for their for their products right after they're used,
15:44
Speaker 1: to incentivize companies to have their products be recycled. Right.
15:49
Speaker 1: Is that am I characterizing that law right? And is
15:52
Speaker 1: it relevant to your business?
15:53
Speaker 2: Yes? So I believe you're referring to EPR or extended
15:56
Speaker 2: producer Responsibility laws. For those who may not have heard
16:00
Speaker 2: of EPR, it's essentially this premise that, you know, if
16:04
Speaker 2: our recycling and waste system is supposed to find a
16:07
Speaker 2: way to do something good with all the stuff throwing out,
16:09
Speaker 2: the people making all the stuff that we're throwing out
16:12
Speaker 2: should probably have some skin in the game to make
16:14
Speaker 2: sure that that stuff gets either disposed of or reused properly. Right,
16:18
Speaker 2: And so EPR laws are already in effect throughout Europe,
16:23
Speaker 2: throughout Canada, some other regions, and then they've been passed
16:27
Speaker 2: in a number of states in the US, including California.
16:30
Speaker 2: Now while EPR in the US is still in its infancy.
16:34
Speaker 2: In other words, it's been passed in a number of states,
16:36
Speaker 2: but there's a lot of hairiness to figuring out how
16:38
Speaker 2: to actually implement the system across all the producers selling
16:42
Speaker 2: into a state and all of the recyclers operating in
16:46
Speaker 2: that state. It is, I think a step in the
16:48
Speaker 2: right direction because in a lot of ways it helps
16:51
Speaker 2: to create that circle we were talking about earlier. You know,
16:55
Speaker 2: you're seeing that a lot of brands and producers are
16:58
Speaker 2: starting to take even more of a vested interest in
17:01
Speaker 2: understanding what is happening to all of their packaging, because
17:04
Speaker 2: they know that imminently they're going to need to start
17:06
Speaker 2: proving the sort of end of life outcomes for that
17:09
Speaker 2: packaging in order to you know, one, not be heavily
17:12
Speaker 2: fined and then two maybe even have a right to
17:15
Speaker 2: continue selling into that state.
17:17
Speaker 1: It seems good that Amazon and Kolgay Palmolive are trying
17:21
Speaker 1: to figure out if the things they make that are
17:23
Speaker 1: recyclable are actually being recycled, but it seems like for
17:26
Speaker 1: that sort of thing to happen at a meaningful scale,
17:29
Speaker 1: you would need laws basically, right. I mean, if the
17:31
Speaker 1: companies are just incurring the cost either out of the
17:34
Speaker 1: goodness of their heart or in the hopes of you know,
17:37
Speaker 1: generating goodwill that will lead to higher revenues, those seem
17:41
Speaker 1: like marginal cases. Are the EPR laws such that you
17:45
Speaker 1: think it will become a meaningful part of your business,
17:48
Speaker 1: a meaningful part of the world, that companies will in
17:50
Speaker 1: fact be on the hook to figure it out, or like,
17:52
Speaker 1: what do you think is going to happen?
17:53
Speaker 2: You know, I will say that early indicators are that
17:56
Speaker 2: all of these states are taking it quite seriously. So
17:59
Speaker 2: in addition to requiring a lot of these brands and
18:02
Speaker 2: producers to pay into a massive fund upfront to even
18:06
Speaker 2: just start implementing some of this movement, a lot of
18:09
Speaker 2: these states are also you know, we're seeing that some
18:12
Speaker 2: of the kind of like early deadlines and fines for
18:15
Speaker 2: non compliance are actually being upheld, which I think is
18:18
Speaker 2: a really strong signal to the market. Hey, this is
18:21
Speaker 2: something that needs to get taken seriously. Now to your point,
18:24
Speaker 2: I do think that at the end of the day,
18:27
Speaker 2: whatever flavor this legislation takes. The key to make sure
18:31
Speaker 2: that recycling is still a viable and sustainable value proposition
18:37
Speaker 2: is that there needs to be some sort of an
18:39
Speaker 2: end market for that material, right because let's say these brands,
18:44
Speaker 2: even if they're required to pay billions of dollars into
18:46
Speaker 2: this EPR system, if there's no one on the back
18:51
Speaker 2: end to receive that material that these recycling facilities are sorting,
18:54
Speaker 2: then recycling can't really happen. At the end of the day.
18:57
Speaker 1: Someone needs to buy the bail of.
19:00
Speaker 2: Plastic exactly exactly. But if they can have guarantee that
19:05
Speaker 2: there is a buyer on the other side, right that
19:09
Speaker 2: that person or that company will buy at a certain price,
19:12
Speaker 2: then suddenly they can sustain that business quite well for
19:15
Speaker 2: the long run. And so to that point, you know,
19:18
Speaker 2: one other model that's often brought up in the realm
19:20
Speaker 2: of legislation is actually minimum recycled content laws, because it
19:25
Speaker 2: kind of gets at the same issue from the other side,
19:28
Speaker 2: where you say.
19:28
Speaker 1: Basically creating demand, creating demand for a bale of recycled
19:32
Speaker 1: plastic coming out of the recycling facility.
19:35
Speaker 2: Exactly, and it kind of disentangles the market for recycled
19:39
Speaker 2: feedstock from the market for virgin feedstock, which is another
19:43
Speaker 2: great way to kind of catalyze the movement of that
19:46
Speaker 2: material throughout that recycling ecosystem.
19:52
Speaker 1: We'll be back in just a minute. What are you
20:03
Speaker 1: trying to figure out right now? What's a big thing
20:05
Speaker 1: you're trying to figure out?
20:07
Speaker 2: We are at a really exciting inflection point, a glacier
20:09
Speaker 2: because of I think two big things here. The first
20:14
Speaker 2: is just how do we scale smoothly and rapidly. You know,
20:19
Speaker 2: we've gone from a year ago we were making maybe
20:22
Speaker 2: one robot every three months, and now we have the
20:25
Speaker 2: capacity to make three to four robots per month, and
20:28
Speaker 2: we're expecting to go even faster by the end of
20:30
Speaker 2: this year in the next six months. And then the
20:32
Speaker 2: other big frontier for us, in addition to just you know,
20:35
Speaker 2: how do we scale and get more of our stuff
20:37
Speaker 2: out there, is what the heck do we do with
20:39
Speaker 2: all of this data? Right? We've already seen that the
20:42
Speaker 2: early use cases for this information people are taking to
20:47
Speaker 2: really in droves. But there's a much more built out
20:51
Speaker 2: version of the data platform where we say, you know,
20:53
Speaker 2: we don't just have a camera in one or two
20:55
Speaker 2: points throughout your facility. We can actually get sensors to
20:58
Speaker 2: sort of blanket the facility, and in that regard, we
21:01
Speaker 2: can take a huge step towards becoming more like that
21:04
Speaker 2: manufacturer who has information on every single step of their
21:08
Speaker 2: process and respond in real time and know exactly what's
21:11
Speaker 2: going on at each stage.
21:13
Speaker 1: So, I mean, let's just talk about the recycling business
21:15
Speaker 1: for a minute. The facilities you're talking about, they're just
21:17
Speaker 1: private companies.
21:18
Speaker 2: The vast majority of them are. So this is a
21:21
Speaker 2: common misconception about recycling, is that you know, these are
21:24
Speaker 2: all you know somehow like public entities buy our estimation
21:27
Speaker 2: about eighty or eighty five percent of these recycling facilities
21:31
Speaker 2: are privately owned, and they can be anything from a
21:34
Speaker 2: family owned business all the way up to the massive
21:38
Speaker 2: waste companies like Waste Management, Republic Services, Waste Connections. These
21:41
Speaker 2: are publicly traded companies that also own many of these
21:44
Speaker 2: recycling plants as well.
21:46
Speaker 1: And the recycling plants are buying recycling from municipalities like
21:52
Speaker 1: they do they pay for whatever tans and plastic jugs.
21:57
Speaker 2: It's actually a very interesting question. It depends a lot
22:01
Speaker 2: on the condition of those end markets. We talked about
22:03
Speaker 2: so in today's climate where those markets are really volatile
22:07
Speaker 2: and a little bit uncertain, oftentimes, you know, these recycling
22:11
Speaker 2: facilities will get paid by municipalities in order to take
22:14
Speaker 2: any process that material. But what's interesting is, you know,
22:17
Speaker 2: back during the heyday of recycling, when you know, China
22:21
Speaker 2: was buying everything, there was no shortage of you know,
22:24
Speaker 2: appetite for that material. The equation flipped.
22:27
Speaker 1: My sense is some recycling is quite efficient and a
22:30
Speaker 1: good business, and some is not very efficient in a
22:33
Speaker 1: bad business. Right, give me the like stack ranking for
22:35
Speaker 1: recycling best thing to recycle to worst.
22:38
Speaker 2: Yeah, the I mean, at the end of the day,
22:40
Speaker 2: the best things to recycle, according to a recycling facility operator,
22:44
Speaker 2: would be the things that most reliably will make you
22:46
Speaker 2: the most money. So top of the stack would be
22:49
Speaker 2: aluminum cans, because there's always a market for those. They're
22:52
Speaker 2: super easily recyclable, and to recycle an aluminum can actually
22:56
Speaker 2: uses about ninety five percent less energy than to make
22:59
Speaker 2: that aluminum can from that virgin or And this gets
23:03
Speaker 2: back to the point about the sort of cost spread
23:06
Speaker 2: between recycled versus virgin feed stuff. Right, the harder and
23:10
Speaker 2: more costly it is to make it virgin, the more
23:13
Speaker 2: of a willing market there is for that recycled material.
23:16
Speaker 1: So cans are great, they always work as a business.
23:18
Speaker 1: Aluminum cans are good. Okay, what's next.
23:20
Speaker 2: Aluminum cans are great. Next is we're going to get
23:23
Speaker 2: a little technical here. HDPE natural. So this is HDPE
23:27
Speaker 2: is a type of plastic resin. If you look at
23:29
Speaker 2: the little Chasing Arrows recycling logo, it's resin number two
23:33
Speaker 2: and most commonly takes the form of milk jugs. Right,
23:36
Speaker 2: that's sort of translucent white.
23:38
Speaker 1: The gallon milk jug exactly exactly.
23:41
Speaker 2: Yeah, and then from there, you know, I'd say it's
23:43
Speaker 2: probably pet bottles. So that's triangle number one. That's like
23:48
Speaker 2: your water bottles, your soda bottles. This is actually a
23:51
Speaker 2: type of resin where we forecast a huge gap in
23:55
Speaker 2: the supply versus what's going to be demanded about five
23:58
Speaker 2: years from now. So that's a really interesting interesting one
24:00
Speaker 2: to watch.
24:01
Speaker 1: And then why I can imagine demand going up, But
24:04
Speaker 1: why can't they just make more of them from virgin petroleum.
24:09
Speaker 2: Yeah. It's a combination of legislative requirements around minimums recycled
24:13
Speaker 2: content combined with sort of like the nature of the
24:17
Speaker 2: end markets that are demanding pet. So you know, pet
24:22
Speaker 2: could be used by the you know, water beverage bottle manufacturers,
24:26
Speaker 2: but a lot of that pet also gets absorbed into
24:29
Speaker 2: markets you wouldn't imagine, like carpet or mattresses or other textiles.
24:34
Speaker 1: So the bad news is that there's plastic everywhere. But
24:37
Speaker 1: the good news is at least they can use their
24:39
Speaker 1: recycled bottles.
24:40
Speaker 2: Exactly exactly, so that that's one turf that's getting pretty heated.
24:45
Speaker 2: And then to at least a round out the sort
24:47
Speaker 2: of container side of things. The other very common thing
24:50
Speaker 2: that gets sorted is HGPE color. So h again it's
24:53
Speaker 2: triangle number two, but it's it's been dyed, right. So
24:56
Speaker 2: this is typically things like your shampoo bottles or your
24:59
Speaker 2: laundry detergent.
25:00
Speaker 1: Jumps, and so is that also like pretty good? Are
25:02
Speaker 1: we still at pretty good on the stack? That's all.
25:05
Speaker 2: That's all pretty good. I'd say those every single recycling
25:08
Speaker 2: facility you go to will sort out those commodities. Okay,
25:11
Speaker 2: And I'll mention here that there's a big time honorable
25:14
Speaker 2: mention for cardboard and for paper. I was focusing on containers,
25:18
Speaker 2: but a lot of people talk about plastics a ton.
25:21
Speaker 2: The majority of the recycling stream is still paper and cardboard, right,
25:25
Speaker 2: so that stuff like that's almost table stakes for a
25:27
Speaker 2: recycling facility. Just you have to get that right if
25:29
Speaker 2: you want to stay profitable.
25:31
Speaker 1: And that's a reasonable business as well.
25:33
Speaker 2: Yes, that's a very reasonable business. A lot of these
25:36
Speaker 2: recycling facilities actually talk about something called the Amazon effect.
25:39
Speaker 2: In other words, as you know, e commerce and shipping
25:42
Speaker 2: has become the way we buy things. Cardboard has just
25:46
Speaker 2: inundated the recycling stream, which is great because you can
25:49
Speaker 2: always sell cardboard now there's a really hot market, but
25:52
Speaker 2: also not so great because maybe your facility was built
25:55
Speaker 2: ten to twenty years ago before this became a thing,
25:57
Speaker 2: and now you have to find a way to sort
25:59
Speaker 2: of jerry rig it to handle all of these massive,
26:01
Speaker 2: oversized boxes that are coming through your stream.
26:03
Speaker 1: So what that we recycle is dumb, isn't really getting recycled,
26:07
Speaker 1: doesn't make sense whatever.
26:09
Speaker 2: Yeah, this is a very long tail of things.
26:11
Speaker 1: You know I mentioned for everything else. Yeah, it is.
26:14
Speaker 2: The everything else. And this points me to what I
26:17
Speaker 2: often tell people is a misconception. But recycling is the
26:20
Speaker 2: phenomenon of wish cycling and I was guilty of this
26:23
Speaker 2: too until I started Glacier and learn a bit more,
26:25
Speaker 2: which is this idea that if you're not sure if
26:27
Speaker 2: something is recyclable, a lot of people who want to
26:29
Speaker 2: do good for the planet are like, I'll toss it
26:30
Speaker 2: into the recycling bin in case they can do something
26:33
Speaker 2: with it. And in fact, this ends up being a
26:35
Speaker 2: huge problem for these facilities because most of the time
26:38
Speaker 2: they can't do something with it, much as we wish
26:40
Speaker 2: they could. So a lot of these contaminants that people
26:42
Speaker 2: throw in there are things like those plastic bags or
26:47
Speaker 2: you know, those films and flexibles, which some facilities can recycle,
26:50
Speaker 2: but most can't. It's things that have plastic in them,
26:55
Speaker 2: but it's not kind of standard plastic. So for example,
26:59
Speaker 2: one very confusing and insidious example that gets brought up
27:03
Speaker 2: often is children's toys. Right, maybe they're made out of
27:06
Speaker 2: some bulky plastic. You're like, hopefully this can be recycled,
27:09
Speaker 2: but chances are that toy has various different grades of
27:11
Speaker 2: plastic on it that aren't easy to pull apart. And
27:14
Speaker 2: heaven forbid it's an electronic toy with the batteries still
27:17
Speaker 2: in it, because that can literally cause an explosion or
27:20
Speaker 2: a fire and blow up the facility.
27:26
Speaker 1: We'll be back in a minute with the lightning round.
27:37
Speaker 1: Let's do a lightning round. It's going to be a
27:39
Speaker 1: little more random. So I know you were a consultant.
27:43
Speaker 1: Is it right that you were a management consultant?
27:45
Speaker 2: I was a management consultant.
27:47
Speaker 1: I'm curious you know now you run a company, right,
27:51
Speaker 1: I'm curious what do you know now from running a
27:54
Speaker 1: company that you wish you knew when you were you know,
27:57
Speaker 1: telling people how to run their company.
28:00
Speaker 2: Yeah. I think a lot of what I've learned running
28:04
Speaker 2: Glacier has been around always identifying and then sort of
28:10
Speaker 2: like revalidating what is that north star metric or objective
28:14
Speaker 2: that we're aiming for, and then making sure that anything
28:18
Speaker 2: else we're working on or anything we're communicating is in
28:21
Speaker 2: support of that.
28:22
Speaker 1: So, like, you're still a consultant.
28:26
Speaker 2: Absolutely, I mean, honestly, I think, like running Glacier, A
28:29
Speaker 2: lot of people think that the tech is the hard part,
28:31
Speaker 2: and don't get me wrong, it's insanely hard, insanely challenging.
28:34
Speaker 2: But when you think about something as cross functional as
28:37
Speaker 2: a circular economy, like I'd say, the majority of my
28:41
Speaker 2: day gets spent thinking about how to align incentives right,
28:45
Speaker 2: like recycling facilities, brands, manufacturers, local legislators, Like they all
28:50
Speaker 2: kind of want different things, So how do you explain
28:53
Speaker 2: initiatives or proposals to each of those parties in a
28:56
Speaker 2: way that makes sense to them and gets everyone growing
28:59
Speaker 2: in the same direction.
29:00
Speaker 1: So if the answer nothing is the answer, you feel
29:02
Speaker 1: like you're actually still doing what you were doing.
29:04
Speaker 2: No, I mean I would say that, you know, one
29:07
Speaker 2: big mindset shift for me that has been very healthy
29:10
Speaker 2: is I'm definitely a perfectionist and a type a personality.
29:14
Speaker 1: You know, by.
29:14
Speaker 2: Upbringing and in management consulting, you're really encouraged to lean
29:19
Speaker 2: into that, right, Like people are paying you big bucks
29:22
Speaker 2: to make sure that you got every every single last
29:25
Speaker 2: detail down to the decimal place right everywhere. And so
29:30
Speaker 2: you know, I'd say that my consulting days were great
29:33
Speaker 2: for training me on like how to make sure I
29:35
Speaker 2: knew what details mattered and really like make sure that
29:38
Speaker 2: everything lined up. But with an early stage startup, it's
29:42
Speaker 2: the opposite where like you don't have time to be
29:45
Speaker 2: perfecting everything, and so that has actually allowed me to
29:49
Speaker 2: sort of flex towards how quickly can I move and
29:53
Speaker 2: still be efficient, Right, Like, what is the right sort
29:55
Speaker 2: of balance of making sure that you're putting out high
29:59
Speaker 2: quality work and that things are generally moving the right direction,
30:01
Speaker 2: but also realizing that actually it's okay and probably good
30:05
Speaker 2: that certain balls are getting dropped, because, as one of
30:07
Speaker 2: my mentors told me, if you find that you are
30:10
Speaker 2: doing everything perfectly and nothing is failing, you're probably not
30:14
Speaker 2: moving fast enough.
30:16
Speaker 1: Yeah. I interviewed the guy who started planning at the
30:18
Speaker 1: satellite company, and he told me that he was upset
30:22
Speaker 1: when none of their satellites were failing. It meant they
30:25
Speaker 1: weren't launching soon enough, they were spending too long to
30:28
Speaker 1: work on it. It's the same, That's exactly right.
30:30
Speaker 2: That's been a massive learning and frankly a pretty painful
30:33
Speaker 2: one in the early years of Glacier, when all I
30:35
Speaker 2: wanted was to make sure that every single thing I
30:37
Speaker 2: outputed was going to work. And you know, at the
30:40
Speaker 2: end of the day, I was like, I just got
30:42
Speaker 2: to get get rid of some of those sort of
30:43
Speaker 2: controlling tendencies if I really want this company to scale
30:46
Speaker 2: at the rate that it needs to.
30:47
Speaker 1: What's one tip to stop being to type A when
30:52
Speaker 1: you're running a startup.
30:53
Speaker 2: Honestly, I don't know if this is healthy, but my
30:56
Speaker 2: approach was to kind of just like overwhelm myself.
30:58
Speaker 1: You give yourself too many things to do so that
31:00
Speaker 1: you have to just pass them on before you're done
31:03
Speaker 1: with them.
31:03
Speaker 2: Yeah, and I'd say it wasn't our intentional per se,
31:07
Speaker 2: because it's certainly not a very pleasant ex experience to
31:09
Speaker 2: go through. But I often joke that, you know, I
31:14
Speaker 2: think starting an early stage company was maybe the only
31:18
Speaker 2: thing that could have broken me of some of these
31:20
Speaker 2: perfectionist habits, because I really had to go through sort
31:24
Speaker 2: of the dark side of pulling all nighters, working myself
31:28
Speaker 2: to the bone, realizing, you know, like what is this
31:31
Speaker 2: all for, and having that sort of existential crisis moment
31:34
Speaker 2: to say, Okay, I don't want to give up on Glacier,
31:37
Speaker 2: and I know we've got an immense amount of potential
31:39
Speaker 2: ahead of us, so I now need to fundamentally rethink
31:42
Speaker 2: how I'm balancing this list of a thousand priorities if
31:46
Speaker 2: I want to do it and still be around in
31:48
Speaker 2: a successful leader years from now.
31:51
Speaker 1: Are you less of a perfectionist in your non work
31:53
Speaker 1: life now than you used to be?
31:55
Speaker 2: Uh? Absolutely? It's like amazing what perspective gives.
31:59
Speaker 1: You on things.
32:00
Speaker 2: Just a lack of time, yes, yeah, yeah, some would
32:03
Speaker 2: call it just a raw lack of time. I do
32:06
Speaker 2: think that it really sort of for you to think
32:09
Speaker 2: much bigger picture about what matters to you and make
32:13
Speaker 2: sure that you're carving time out for that and then
32:16
Speaker 2: just not sweating the details. And the amazing thing is
32:19
Speaker 2: once you start doing that and you realize that the
32:21
Speaker 2: world isn't going to end because you forgot to do
32:24
Speaker 2: this thing or decided not to do that thing perfectly,
32:26
Speaker 2: it gets easier and easier to do right. So that's
32:29
Speaker 2: been really healthy for me.
32:36
Speaker 1: Rebecca who Trums, is the co founder and CEO of Glacier.
32:41
Speaker 1: Please email us at problem at pushkin dot fm. We
32:44
Speaker 1: are always looking for new guests for the show. Today's
32:48
Speaker 1: show was produced by Trinamnino and Gabriel Hunter Chang. It
32:52
Speaker 1: was edited by Alexander Garreton and engineered by Sarah Briguerra.
32:56
Speaker 1: I'm Jacob Goldstein and we'll be back next week with
32:59
Speaker 1: another episode of What's Your Problem