Nathan Shammah, CTO of the Unitary Fund, is interviewed by Yuval Boger. Nathan discusses the non-profit’s evolution from focusing on software to hardware in quantum technologies, their Mitiq project for error mitigation in quantum computing, and the fund’s success in establishing a new research institute and promoting open-source quantum technology. Nathan also muses on a potential breakthrough moment for quantum technology, discusses open-source hardware projects, and much more.
Full Transcript
Yuval Boger: Hello, Nathan. Thank you for joining me today.
Nathan Shammah: You’re welcome. Thank you for having me.
Yuval: So, who are you, and what do you do?
Nathan: Yeah, I’m Nathan Shammah. I work at the Unitary Fund, where I’m Chief Technology Officer. And at Unitary Fund, we support an open ecosystem for quantum technologies. We’re a non-profit, and we believe that the most efficient and equitable way to achieve this is by supporting open source projects.
Yuval: What kind of open-source projects? Is it hardware or software, and if it is one of them, could you give me an example of some of the projects that you’re supporting today?
Nathan: Yeah, that is a great question, actually, because it really started the focus on software, and we recently shifted to also incorporate hardware. So maybe I can take a couple of minutes just to give a bit of a history of how the Unitary Fund started. It was around 2018, I don’t know the exact days, but that’s when Will Zeng started it. It was really an experiment of giving fast grants to explorers in the space. So, one way of having an impact in this space where there is so much to build is through software because there is so much software that needs to be built as an infrastructure. And starting from then, when the vision and the mission expanded also in staff, and now we have about 11 folks as part of Unitary Fund staff, also working not only on funding projects that explorers in the space get funded to explore, just as with quick grants, but also with long term projects that we believe we are best positioned to tackle, and I can speak more about that.
Next to this, the focus is still on open source. I would say software, but as a byproduct almost, we also end up doing a lot of research. So that’s also what I and the staff work on, and we have two main focuses at the moment, which are error mitigation and benchmarking. So this is sort of a summary of what’s the Unitary Fund, and maybe to answer more directly your question about software and hardware, we’ve recently written together with a bunch of experts in the field a review, which is entitled “Open Hardware in Quantum Technology”, which you can find on arXiv, which is actually the first review of open source software to control and design hardware, as well as hardware and firmware itself that is open in quantum technologies, including of course quantum computing. We’ve recently given a couple of grants to projects that include, for example, as an example of one of these grants, it’s the OpenQuantum.org project, which is a really fun project by Max Shirokawa Aalto. And basically we’re funding him, and again, this is no-strings attached, it’s all charitable funding, it’s not like startup funding from VCs. It’s really like to develop a blueprint with CAD files for a magneto-optical trap, including electronics, command-atics, control firmware, and assembly instructions. And this is going to be all open source and on OpenQuantum.org.
Yuval: Let’s start with the grants actually, the smaller grants. I think a while back, you had this micro-grant program where I guess if I’m a student and I’m interested in doing something over the summer and improving some open source project, I can submit a short application. Is that still the case?
Nathan: Yes, exactly. And this project that I mentioned is just one example of the most recent grant that’s been awarded or announced, and it is still ongoing. So, about 92 grants have been awarded so far to folks in more than 26 countries worldwide. And as you were mentioning correctly, these are $4,000 grants. It’s very easy to apply. Just go to Unitary.fund, and there is like a really two-minute application. I’m not joking, it’s just two minutes. You cannot do more. Usually when it’s long applications, they’re rejected. Of course, you can just reapply the next day. And some brief descriptions from a typeform. A member is an advisory board that we put together. Actually, I really want to thank the volunteers from this advisory board who helped review these grants on a monthly basis. And yes, usually shortly in a month or so, you’re going to learn if you’re going to win a grant or if you got some constructive feedback. And usually what is really fantastic from my point of view is that many awardees actually had multiple attempts to get a grant. They get this constructive feedback from folks in the field who come from various places, from various continents, institutions, and in corporate and academia, who are really developers in these fields. And they can spot the application, if there’s a real need for something, if there’s some talent, some talented applicant that maybe could focus on something else.
Yuval: How is the Unitary Fund funded? I mean, if you had 100 grants at $5,000, give or take, that’s half a million. And now you have 11 staff members. They must be paid something. Where’s the money coming from, if I may ask?
Nathan: Yeah. So, the grants are 4K. When it started, they were 2K. So dollars, it’s a bit less than that total funding for microgrants. But yes, actually, this is a great question because we started two years ago a member program. So this is a way for organizations to be part of the Unitary Fund. And we’re really happy to be supported by these organizations. You can find them all on unitary.fund. And we have basically two tiers of support. I would say they’re broadly from two or three buckets. So, they are folks who support what we do because they believe in the mission. Well, they all believe in the mission, but some find it useful as a way to explore this technology landscape. And some are already in this landscape. So, we are like some of the startups or corporates that are part of this network. And they really find value in the work we do for many reasons. I mean, they are usually nerds who like to develop software and projects. They like to source talented folks. And they like the sort of programs that we have broadcast.
Yuval: One of the projects you mentioned is the error mitigation one. I think it’s called Mitiq. There are plenty of commercial companies that are working on air correction, error mitigation, and error suppression. What can an open-source project do that a commercial company that is likely better funded cannot do?
Nathan: Yeah. So, yeah, this is a great question. And we believe that there is a use case for an open-source layer for error mitigation that should be available to everyone to use. And that’s the main reason why we started Mitiq. We hope that at some point, Mitiq will run on any quantum program that is launched either on the cloud or locally. And we also believe that there is a lot to create and test. And this is something you see in science, but also in other fields like machine learning that are more mature. You can really leverage open source platforms and communities to build faster and to experiment faster. And that’s basically the main reason why we want Mitiq to be open-source so that everyone can use it.
Yuval: The Unitary Fund, if I remember correctly, has existed for a number of years now. What do you consider the biggest success?
Nathan: I think we’ve established, and when I say we, it’s really like a broad network of folks starting from tech staff, board members, advisors, and everyone who participated in this community… We really established a new kind of research institute that is not something trivial to happen, I think, in science in general. I mean, you see things like the Flat Iron Institute, and that was, you know, Simons was able to fund it. I mean, he didn’t need a member program to make it work. But in this case, I feel like we really started doing something that is making some statements. And these statements are that we have the opportunity to make this technology open. Open source can have a crucial benefit and impact on this technology landscape. And there is a common good, as Will always says, between a venture-funded startup and what you need to focus on for profit, of course, and what academia is currently focused on. So there is a space that includes a lot of infrastructure that can be taken, and this can really be a platform for years to come, in my opinion, for explorers in this space.
Yuval: And on the flip side, professionally speaking, what keeps you up at night?
Nathan: Well, I mean, really keeping me up at night, it’s really a blast. I had so much fun, and I enjoyed so much being at the Unitary Fund since I joined about three and a half years ago. It’s really like an adventure. I think it’s exciting to see how we are evolving as an organization where we’re growing. It’s not like in a startup where we have these like explosive growth. But yeah, it’s slowly growing. And to go back to your question about how the Unitary Fund is funded, I think it’s also really exciting that the Unitary Fund now got awards from three different entities. The Department of Energy, a few years back, and this grant is still ongoing, the Welcome Leap and the National Science Foundation, who found something really nice in what we were proposing, sometimes with partners, with amazing partners, and what we wanted to build. So I think this is also something really interesting to see the diversity of support that such a nonprofit organization can keep attracting in the future years.
Yuval: Where are the members of the Unitary Fund, sort of the 11 or so employees? Are they all in the same area? Are they all over the world? How does that work in practice?
Nathan: Oh yeah, this is what I would say. This is what keeps the members up at night just because we’re so delocalized geographically. But in a good way. So some wake up early, and some go to bed a bit later. Just joking. So we actually were fully remote. And I think this was quite amazing because when really the technical staff started working, it seemed quite a real experiment when, as you know, COVID happened, and many companies switched to remote. And we were already robust from this point of view. So it’s a mix of the United States. Currently, it’s scattered across the East and West coast, the United States, Canada, and Europe. And so it’s really broad.
Yuval: I saw a while ago, and I think I actually filled it as well, a survey that you did on open source. Are there any results that you can share from the survey?
Nathan: Oh yes, this is super exciting. This is the second year we run this survey. Results are just in, and you can find more at Unitary Fund. We also have a blog post with some analysis. So, we were really inspired by the Stack Overflow Developer Survey. Of course, there are questions that are specific to quantum technologies, but also broader questions about developers and users of quantum software overall. And now we’re starting to have a year-on-year comparison. So this is really interesting. One thing that I think is generally super useful to basically everyone in this field is what programming language is most popular. And no doubt Python is still in the lead, about 94-95% year on year. There is, of course, an increase of Julia adopters. That’s also interesting. As well as Rust. Programming language like C++ it’s a bit receding. So this is something quite interesting that I found from the survey.
Yuval: What projects would you like to start that you don’t have the people or the funding to do right now?
Nathan: Well, this is great. I mean, this is a great question because we have many ideas. And one thing is that we actually have an open position for an open-source quantum fellow. We have a sort of postdoc-like position that doesn’t need to be too preoccupied with publishing papers, but they can also do great things for open science by opening pull requests and making tools integrate more. Like location. jl that works with some other front end; with Qiskit, you make sure that things on Braket work with some other backends, and so on and so forth. So, the current position we have is going to be focused on quantum error mitigation and Mitiq ecosystem, extended ecosystem, and integration. I think we should expand this, we should repeat this. And having really the opportunity to attract talented people that can really code for work and work on open science software, that quantum machine, that’s great.
Yuval: If I wanted you guys to work on something, I don’t know, an open source compiler that converts this to that, is there a way for me to suggest a project to you, or how do projects get started at the unitary fund?
Nathan: Oh, definitely. I mean, of course, we could start; we’re really interested into compilers, and one way of looking at Mitiq is as a compiler. But yeah, this is, I think, a very interesting space in particular. In general, with collaboration and projects, this is something we’ve done in the past. And yeah, you can just send us an email, I mean, nathan@unitary.fund. Of course, we also have some more general contact points like info@unitary.fund. We have a lot of online meetings where people can just dive in or propose to give a talk or start a discussion. We have a Unitary Fund Discord server with thousands of participants, and we have many weekly calls, including one that is quantum Wednesday, which basically every time we meet, and it is really broadly from a journal club to a discussion, a very informal discussion with folks online. Also, of course, people can reach out directly.
Yuval: I have two more questions before we finish today. One is, what do you think will be the chatGPT moment of quantum? What do you think it will be, and how soon?
Nathan: That’s a great question. Hard to answer. You know, the chatGPT moment, from my point of view, happened because I put it easily. It’s not as easy, but technology basically got commoditized. Or at least it’s clear that with direct effort, one could take some leap, let’s call it a quantum leap in some direction, at least from a product perspective. In quantum, I think we may be there. You can always have this quantum leap, but of course, we need to improve. There could be some emergent phenomenon, there could be a new architecture. I think almost everyone agrees it’s basically an engineering problem. Maybe it’s several orders of magnitude of distance. And then, when we get closer, I think we’ll be closer to a chatGPT moment.
Yuval: And last, hypothetical, if you could have dinner with one of the quantum greats, dead or alive, who would that person be?
Nathan: Oh, this is great. I’ll think about it in a second. I think, well, this is tough because, you know, I could really say Einstein, but let me not be so trivial. Let me say John Stuart Bell from the Bell Inequalities. And yes. Okay, if I can cheat, I mean, he’s not dead as far as I know. It’s also David Deutsch, someone who I really admire who would be great to speak with. So yeah, but let’s stick to Bell.
Yuval: Very good.
Nathan, thank you so much for joining me today.
Nathan: It was a pleasure. Thank you