Afternoon, and thanks for joining us to have a conversation with Jeff Hawkins, President and CEO of Quantum-Si. Quantum-Si is a proteomics technology company redefining protein analysis through single molecule protein sequencing with a benchtop platform that brings real-time kinetic-based protein analysis to every lab. The company's first generation Platinum system is in the market today and is being used both to generate evidence and to inform the design of the next generation system, Proteus, which is being planned for a commercial launch in 2027. To talk about the experience with the Platinum and the launching of Proteus, I welcome Jeff Hawkins to this presentation. Thank you. Jeff, thank you very much for being here. To start off, you have stated that Proteus launch is targeted in the second quarter of 2027 and Platinum has been in the market, which has been generating evidence. How would you frame the company's strategy, let's say, over the next 12 to 18 months, and what does success look like by the time we get to the launch? Yeah. I think success over the next 12 to 18 months really is underpinned mostly by the completion of the development of the Proteus platform, and the subsequent commercial launch, and then obviously the uptake on the backside of that. Platinum, as you mentioned in the opening remarks, has really been used, especially in the last sort of 24 months, as a market development tool. I think it continues to provide us opportunities to get people an early sort of flavor of the technology in their lab, as a way to sort of onboard them into protein sequencing and ultimately into what we think we'll deliver with Proteus. I think success is that launch, success is commercial adoption. Then I think what else could sort of come from that, partnerships and other things we may be able to do with that platform in the market would be sort of the additional upside in that timeframe. Perfect. Single molecule protein sequencing is not something which is in everybody's vocabulary. Can you explain what that means and what does Proteus do, what mass spectrometry or immunoassays have been used for in terms of proteomics? Sure. Proteomics is not new. It's been around for decades, predominantly through things called immunoassays, right? You have an antibody that you want to bind to a particular substrate, maybe an antigen to bind to a complex, typically looking for just a parent protein or single sort of thing to detect. Mass spec has been, I think, really should be credited with a lot of the growth in proteomics over the last, call it 10 to 20 years. There, you chop the protein up into smaller pieces. Those are thrown inside a mass spec, and you measure the mass, and you can sort of say something about the individual small pieces called peptides. Single molecule protein sequencing means literally those same peptides, but they're bound into a well, and you're able to detect, one at a time, each individual amino acid. You can see that sequence in order, you can see small modifications to that sequence in order, much more akin to what many of us know as DNA sequencing, except in this case, the alphabet's 20 letters and you're reading the sequence of a protein. What people are excited about with the technology, one is getting a complete sequence of a protein, rather than sort of inferring that from fragments. The ability to measure those small modifications called post-translational modifications. With sequencing, the concept of being able to work reference-free, so without having to know what's in the sample. Mass spec is a great tool, but you need to know what that reference database is. What do you think is in that sample? I think those are maybe broad strokes, the things people are really interested in doing with sequencing that's not necessarily accessible today with current technology. One of the things that you recently have started talking about is you're very clear that the move to the delay in launch to the second quarter of 2027 was internally driven decision, and it's not a consumer feedback or a shortfall of the sequencing chemistry itself. Can you walk us through what the integrated units actually have showed you in the first half of the year that prompted you to make that decision? Sure. If we take one step back, in 2025, we had built some prototypes of the machine, and those prototypes are not complete in terms of having every single component of the final instrument integrated. Based on those prototypes, we had shown some sequencing data at our Analyst Day in November of 2025 that showed that our sequencing chemistry that was on the first gen technology could be ported over, worked effectively on the prototypes. We then went into an integrated cycle, so bring all of that together with the liquid handling automation. The first thing to say is, it sequences. We don't have any machines that have ever shown up that just don't work. We don't have that sort of what I would call a fundamental technology challenge. What we saw in the integrated units was, some instruments came in and have performed at the level we would expect in terms of various metrics we look at. Others, while they were sequencing well, they weren't at that same level. They were a little bit lower. Again, none of them were not functioning, and we weren't seeing wild variation in machines. It really felt to us like this was about just improving the repeatability. It was about getting the right specifications and tests in place in manufacturing to ensure that every unit coming off the line performs the same way. When you look at these things, there are two options that we considered. One option is go from the current version straight to what people might call a pre-production or a production unit, depending on the lingo people use, or add another cycle in there, digest all of the changes and improvements in the end-of-line testing, get confidence in that, and then move to the production stage. We weighed those two and decided the extra cycle made the most sense. It de-risked things. It gives us higher confidence that before we take on that big expense of going into production, everything is working the way we want. In parallel, though, I mean, you're still working with those existing units, continuing to improve the performance through other aspects of the technology, including the chemistry and the software and algorithms. We're definitely not stuck. We're just adding that cycle to de-risk the step rather than jumping straight to production. You also mentioned instrument-to-instrument variability with some units performing above the internal specifications, whereas some below. Adding this additional design cycle, how is that going to help out? And what sort of, I am not asking for all the details. Sure. But at a higher level, what are the things that you need to tweak so that by the end of that design cycle, you are confident that you have a fully functional instrument? Yeah. I think if we look at variability, one portion of variability is sort of how the sample gets prepared and loaded into the device. That is something we can work on right now, and we have actually seen some really promising improvements in that sort of pre-instrument aspect of the workflow, even over the last month or so since our last earnings call. Then you have more the instrument component, and there it is really about making sure that the test we run, whether that is on an optics module or whether that is on the finished instrument, we want to get the specifications of that. W hat equals passing or failing, really fine-tuning those and linking them back to that high-performing sequencing. If we get that correct, then that assures us that those units coming in are always coming in at the higher level of performance. Marry that up with the improvements we are making pre-instrument, and we think those two together really lift the performance robustly, deliver sort of instruments. Because again, this is not about proving we can build 5 or 10 or 15 of these. It is about having a process to be able to make 50, 100, 200 of these to support sort of a commercial ramp. Okay. You also have made some changes in the leadership and also in the governance. Changes have been put in for explicit product readiness. Can you highlight to us some of those changes? Are there any of those that you feel confident that you will get through in terms of getting the launch done? I mean, I think the main thing we did is we really sort of flattened the organization. Today, the way we have it structured, all of the key R&D disciplines, whether it is algorithm science or recognizer development or engineering, all of those functions are reporting flat into me. I am taking that accountability for stitching together those activities, ensuring we really have that product point of view. I think similarly, we have increased sort of some of the resourcing and support from some of our development partners. That gives us both the benefit of being able to surge resources and capture their expertise, but also turn off that spend later when we launch. I think those are a couple of the major components. Obviously with it reporting to me, there is not any sort of risk of information flow or the team having the right direction or the wrong information coming to myself or our CFO for decision-making and the board. I think sometimes flatter, it creates some additional work for yourself personally, but it just leads to more efficient communication upward and downward within the organization. One last question on this part of the platinum went through this one time, right? I was first introduced to the company just before that. I really commended what you did there in terms of pulling the Platinum out from the market, working on it, and making sure that the right and functional device was put back in. What were some of the learnings there that you felt like you don't want to go through that process again, that made you make this decision? Well, I think what I was focused on is what was the performance we were seeing? What were our estimated cycle times to make a machine, deliver it, and have it running? I was looking at the various metrics that at any given time in a program should be improving towards what you'd expect at launch. You don't expect mature product life cycle level efficiency, but you do expect to see each of those metrics moving towards where you want. I didn't see that happening in sort of that wave. I saw certain areas advancing and other areas perhaps not moving forward as quickly. I think we learned with Platinum that this is complex. Sequencing proteins is not simple. There's a reason why we're the only ones who have commercialized even a first-generation technology. This is not simple. It's not just about engineering, the biochemistry, the analysis, all of these things have to work together. I think we learn from Platinum, you can't be over-indexed to just one thing. Don't just focus on the machine or just on the chemistry. You really do have to look at it as a product. That's sort of what I was watching and didn't quite see that coming together the way we liked. Try to be proactive whenever possible. You don't want to jump too quick to a decision. They're not simple decisions. They can be disruptive if you're not careful. S aw enough data to know we needed to do a few things differently to get the outcome we're looking for. Talking about wins, this morning you did make an announcement about a win, which is that you can detect 18 amino acids now. You also stated that by the time you launch, you would try to get to 20, which was not the statement that you had made previously- Correct With the previous timeline. Number one, on the 18 amino acids, what does that help in terms of sequencing itself? Then, what has happened in the interim that gives you confidence that there is a good possibility that you could get to 20? L et's maybe start with why does it matter? There are certain applications in the market, applications like antibody sequencing, the desire to look at mixes of proteins where there are unknowns in there, and therefore higher coverage and higher knowledge of the sequence is helpful. Those areas really benefit from getting as close to complete coverage as possible. Now there's a whole list of applications, and there's been plenty of publications with the technology that show you don't necessarily need all 20, but those are some key applications that would be unlocked. Then there's just the reality of customer buying behavior. I think if you've been around DNA sequencing, you say sequencing, it means the four letters in DNA. Yep. Similarly, in protein, try to convince somebody that you need less than 20 when the alphabet is 20, and some people just get stuck. I think what they're thinking is, "If I'm going to make the leap to a new tech, I want to use it when it's done." I think there's both aspects of why we're pursuing 20. I think in terms of what's happened, so really where we're at today has a lot to do with a big move we made almost a year ago now, which was to bring in an AI tool that we could train on a local instance of all of our data around developing these small protein domains that can bind amino acids. Without getting too technical, we were using all of the tools everyone knows, AlphaFold, AlphaFold2, you name it, we were using those. T here's not a lot of information in the public domain about what makes a protein bind an amino acid. We have that. We've been doing this for almost 10 years, if you go all the way back to the start. Training the model on our data has radically improved our ability to both identify a new recognizer and improve its characteristics to work in sequencing. I think t hat the scale-up in the throughput of the activity, and then with the instrument timeline moving out a bit, it gave us a little longer to complete the reagent work before it had to transfer. I think you put those three things together and we feel like we're in a good spot to try to get to that 20 by the launch. Very good. In terms of commercial readiness, you have been working on this for quite some time now especially through doing road shows. You have done road shows not only in the U.S., but also in Europe. If you can highlight what these road shows are meant for, and when you look at the European market and the U.S. market, what are the differences at a high level in terms of the appetite and also the mix of potential customers? Yeah. First of all, a road show is an intentional decision by us as this is the better way to reach customers than some of the classic marketing methods. Going to a trade show and having a big presence and a booth and all these things. We go, but we go in a very small capacity. We find it is difficult to get a lot of attention and a lot of time from customers. A road show is literally, we go to a site. That could be a pharmaceutical company, it could be an academic center, sometimes it is an offsite venue that people come to. But customers, potential users, come there. We have presentations on published evidence around the technology. We give information about Proteus. Sometimes there is a customer speaker who will speak about having used Platinum. It is not at every one of the road shows, but some. But what it gives us is a focused 90 minutes to 2 hours with a fairly large group of users in an institute. And we have just found that that has been far more effective at both educating but also learning where the technology could be applied, who is interested in the institute, and starting to get a feel for, with multiple people interested, who might be the actual acquirer and operator of the platform versus I want to provide samples to that person to do the work for me. So it has been a very effective tool. We see a lot of the same information from U.S. versus European. The one difference I would maybe call out in Europe is here in the U.S. you have these massive genome centers, and in a similar way, you have these really consolidated, huge proteomics cores, just filled with top-end equipment. That's not as common in Europe, and in some instances, we're seeing more that the genomics lab is trying to become multi-omic. Because you might not have that big proteomic core lab that you have here in the States. It's not universal, but we see a bit more genomic becoming multi-omic interest in our European road shows than we necessarily see here in the U.S., where proteomics people seem to be a bit more focused, and genomics people are a bit more focused. It's not that they don't talk about multi-omic, but different people have different technologies for different purposes. I know it's a bit early, but you certainly have put out the price for this, for the Proteus. When you think about, based on your experience with Platinum, where do you think which bucket of customers you think are a low-hanging fruit here? Is it academic, industry, government? Well, the first thing I'd say is when it comes to price, if your technology is solving a problem, the pricing component tends to be less impactful than I think a lot of times people think it is. There are million-dollar DNA sequencers being sold every day because they solve a lot of problems. Yeah Million plus dollar mass specs also being sold because they're a very powerful tool. I think delivering the capability, the coverage of amino acids, the ability to look at PTMs, these are the things that will dictate if the price point of the machine is correct. I think if we have those capabilities, I don't have a concern about the price. In terms of where, I think new technology like this often is a little bit overweighted to academic. I think there are clear opportunities in pharma. I expect them to move a little slower than academia. That's pretty typical. They have a lot of capabilities already, so they're going to be more scrutinizing and making sure you're going to solve their specific problem that isn't being solved. I think there's publications on this, and given our existing traction, especially on the defense side, I think that is a potentially interesting opportunity for us. They've been applying our technology for a while. They've been publishing on our technology in the biothreat space. We continue to interact with them, across multiple branches and even some of the militaries outside of the U.S. I still think that is a good opportunity. It's a little hard to tell the exact time they might move, but we feel good about, if there's a movement to adopt a technology like this for biothreat, we're in a really good position to do it. Okay. You're doing the road shows now. You're introducing the instrument now, but at the same time, you have to tell them that you've got to wait for a year to have this in your lab. How do you keep that funnel warm and still interested over the next four to five quarters from here? Yeah. The first thing I'd say is customers have seen novel technologies take a bit longer than the companies expected. We're not the first to have that type of communication. I think most customers that I've interacted with since the announcement weren't overly concerned. Six months difference doesn't sound like a lot to them. The rationale for why was not concerning to them in terms of, we didn't run into a fundamental technology problem. Got it. But I think we'll continue to do road shows. I think the next thing that would really help would be if we begin to allow customers to send us some samples to generate some data. Then ultimately, when we place these in end users a few months before the launch, that's when a lot of customers will go, "Okay, now it's coming." So I think it's our job to keep them informed but not push too hard when we're not quite ready to deliver the platform to them. So it's a balance, but we've got really experienced salespeople. We have a great leader of our commercial team. I think we can strike that balance. Okay. In the recent reduction of force that you did, you deliberately made sure that you protect your high-priority Proteus work streams. With that, how confident are you about the commercial capacity itself, and is it still you feel comfortable enough that you have all the right people to launch in the second quarter of 2027? Yeah. Listen, I said this earlier. Sometimes you have to make these difficult decisions on how to allocate capital. I think the decision we made to reduce the total size of the workforce allowed us to extend that cash runway out into the fourth quarter of 2028. That gives us that solid 18 months post-launch. More than two years from today, but that 18 months post-launch. We are confident we have got the dollars being invested in the right places commercially. I think our assessment is outside of maybe some fine-tuning, some field service engineering capability to augment what we have. We think we have got really what we need. We have got a good level of coverage. The road shows are getting us into cities where we had less traction or less awareness before, so we feel good about it. I view it more as a fine-tuning to get to the launch in 2027. We really then see the uptake, and if the uptake is the way we expect it to be, then we would see more scaling into 2028, not really in advance of the launch, more in response to the uptake. We have run past our time. In the next 30 seconds- Yes What do you think are the two or three things that people should be looking out for between now and the launch? Well, I think in terms of just are we on track, where are we with the amino acid coverage? We will continue to communicate as we demonstrate the capabilities and as they are integrated. From an instrument perspective, where are we at on this integrated cycle? And ultimately then the placement into external users. All of those, to me, are key events that have to happen to deliver on that launch in time. I think people looking for those milestones. And of course, we are out every day. We are trying to attract early interest. And if we were to be able to get any pre-orders or anything, and be able to announce that would obviously be an enormous milestone to be able to capture that before we are able to launch. I think those are the things I think about. Thank you. Thank you very much, Jeff. Thanks. I appreciate your time. Yep. Thanks a lot. Thanks.