Lab of the Future | Where We Should be Heading with Olga Seltser, M.E.

Welcome to another installment of our series focused on the lab of the future! In our previous episodes, we laid out the foundation of what the lab of the future is and how this initiative can advance research. Now, we're setting our sights on identifying where we should be heading with the lab of the future to get closer to making the concept a reality. 

Our guest is Olga Seltser, M.E., senior director of automation at Laronde & Harbinger Health, an experienced automation engineer with a demonstrated history of working in the biotechnology industry.  Olga shares her perspective on the opportunities to advance laboratory automation to make progress toward the lab of the future.

For a transcript of this episode, please visit this episode's page on Buzzsprout.

Key Learning Points: 

• How automation in the life sciences compares to other industries like the auto industry
• What needs to be done to elevate life sciences automation to the same level as other industries 
• The role higher education plays in the lab of the future
• How to make automation more accessible to labs with limited funding

New from SLAS - the Lab of the Future Short Course on Demand!
SLAS recently launched Lab of the Future, a new short course on demand offering to educate new laboratory technologies and applications to make any laboratory more advanced!

To learn more about the Lab of the Future short course including availability, pricing and the instructors, visit:
https://www.slas.org/education/slas-short-courses/lab-of-the-future/

Stay connected with SLAS

• Online at www.slas.org
• Facebook
• Twitter @SLAS_Org
• LinkedIn
• Instagram @slas_org
• YouTube

About SLAS
SLAS (Society for Laboratory Automation and Screening) is an international professional society of academic, industry and government life sciences researchers and the developers and providers of laboratory automation technology. The SLAS mission is to bring together researchers in academia, industry and government to advance life sciences discovery and technology via education, knowledge exchange and global community building.  For more information about SLAS, visit www.slas.org.

Upcoming Events:

SLAS Meet-Ups

• Leiden, Netherlands (10 September 2026)
• Tübingen, Germany (20 October 2026)

SLAS 2026 Sample Management Symposium (October 21-22, 2026 | South San Francisco, California)

SLAS2027 International Conference & Exhibition (January 30 - February 3, 2027 | San Diego, California)

Thrive in Science: A Women's Leadership Forum 2027 (January 31, 2027 | San Diego, California)

View the full events calendar

Transcript

Copy Transcript Copied

Hannah Rosen: Hello everyone and welcome to New Matter, the SLAS podcast where we interview life science luminaries. Today, we'll be continuing our series, focusing on the lab of the future. Joining us today is Olga Seltzer, Senior Director of Automation at Laronde and Harbinger Health, here to discuss where we should be heading in the lab of the future. Welcome to the podcast, Olga. 

Olga Seltser: Thanks Hannah, I am delighted to be here, and thank you for having me on the podcast. 

Hannah Rosen: It’s our pleasure! So, to start off with, I was hoping that maybe you could just provide us with a little bit of your professional background. 

Olga Seltser: Absolutely. I kind of stumbled into the world of laboratory automation to be honest with you. My undergrad and my grad were in biomedical engineering, and I was aspiring to design minimum invasive neurosurgical equipment, and it didn't turn out that way. I started my first professional job at Philos, a company that hasn't existed in a long time, was put up in front of a Tecan liquid handler and kind of said... was told make it work. And, you know, from there on I never looked back. So, that was the beginning of my laboratory automation career. You know, I've gone through the big Pharma experience, medium sized biotech startups that have gone public, and more recently the last two years I've spent with Laronde and Harbinger Health, you know, flagship pioneering companies, small startups that have had explosive growth over the last two years. So, you know, kind of diverse experience as to the levels of automation, the budgets, the complexity of automation has been all over the place in my last 20 years. 

Hannah Rosen: That's great, so you certainly have a lot of knowledge and personal experience of the wide range of types of labs that are trying to get into this automation space. 

Olga Seltser: And you know, having had that diverse exposure to different organizations, I've also had the opportunity to learn from all kinds of walks of life. You know, people that are scientists by nature that have transitioned into the world of automation, straight engineers that didn't know science and have had to pick it back up, software developers that also transitioned into the world of laboratory automation, and all of that brought very different perspectives, knowledge and skill set. 

Hannah Rosen: So, I'm curious, is that, you know, this is obviously something that is a passion for you, and, you know, in your mind, what discrepancies do you see between where the "lab of the future” is currently heading in life Sciences versus where, in your opinion, it should be going? 

Olga Seltser: That's a great question, and I don't know that, you know, everybody that's listening to this podcast would be on the same page, but at least from... from my perspective, there's probably a couple of things. The pace of digital that supports the automation development, or deployment, I should say, I think that's where it's still lacking. I feel that this area would benefit from a lot more software developers that understand a little bit of the science, but also, how does the automation actually work in the lab? You know, I think there is a little bit of a disconnect there that we see. The other aspect, I think, is the workforce that ends up using these robots. I think we can do, and have to do, a much better job educating our next generations and giving them the tools that are going to make them successful in the labs. Because if you... if you think about it, so many folks in the science labs are PhD scientists, you know, you're talking about somebody that devoted a good portion of their life to a high degree of education, and they would put them in the lab and we asked them to pipette manually. So, there's a couple of issues that I see with them. You know, we... we really want to extract the brain capacity, right? What the scientists can think, what they can design, and all of the innovation that comes with, you know, somebody within that walk of life, and also give them the tools to actually drive the science that they are so passionate about. So, you know, I... I think these are the two major issues that I see. So, you know, the under skilled workforce to bring automation to fruition and also, sort of, the digital component, the software development that needs to walk in stride with automation to be successful. 

Hannah Rosen: I'm curious, you know, in your opinion, for that part of, you know, having people properly trained and skilled to work with these automation technologies, do you feel that it... we’d be better off training the, you know, scientists in the lab who are running the experiments on all this equipment, or are we looking at having a specialized group of individuals who, that is their primary job is to operate and run these automation machines and have that be separate from the other, you know, lab scientists? 

Olga Seltser: Yeah, that's an... that's an interesting perspective. I mean, I would... I would say that you kind of need both. You know, somebody that really understands the hardware, and how to extract the most on the software level from that hardware, but also understands the science. And then the other partner, on the other side, would be a scientist that has, you know, really in-depth knowledge of the biology or, you know, whatever field, could be agriculture, whatever they're working with, but also understanding enough of the capabilities and maybe the... the current drawbacks or challenges on the automation side, because this is where we see, where I see at least a bigger gap. You know, we have the automation people that either have had limited exposure to science, or, you know, just enough to be dangerous. And then we have the scientists that think that automation is as easy as opening up a box and setting up, you know, like an analytical instrument where you create your method. It's straightforward and it just works from the get go, which is typically not the case. So, understanding what's possible, what's reasonable, and what are the possible adjustments on the bench, workflow that might need to be made to allow automation to... to be successful in... in that environment. So yeah, this is... that's a long way to answer your question. I think you're going to need to have both, and have both sides complement each other. I think, you know, it's never too late to learn, so there's probably no right answer. We have now seen a couple of different universities introduce robotics as a program in their graduate programs. I believe the University of Uppsala is one of them, and I think Cornell is another institution that's introducing that into their graduate programs. But to be honest with you, I would love to see this being introduced as early as possible, you know, even at the high school level. We actually just did a quick segment, there's a local chapter of Science Club for Girls that is exposing, you know, the... the 5 - 12-year-olds to STEM education, and we just did a quick video introducing them to robotics in the lab. What those absorb is so different, but the ability to expose the young adults at an early age, just to educate them what's out there. I feel like automation engineering is still a niche and so, so many people don't even know that this avenue is... is something that's available to them. So, I think that the earlier we can do that, the better we serve, you know, the field and the community as a whole. 

Hannah Rosen: Yeah, I think that's a really great point. I mean, speaking of the automation industry as a whole, you know, how does automation in the life sciences compare to automation in other industries, such as the auto industry? 

Olga Seltser: That's a wonderful question. And the reason I say that, I've never thought of it being, sort of, underdeveloped, or immature might be a better way to say it. We were at dinner not too long ago, and the CEO or Co. Founder of Artificial, David Fuller, was there, who comes from, you know, true robotics background. And he... he kind of made a point, or a comment saying, you know, this field in the life sciences is really immature. And I, you know, I had to take a step back and... and explore that topic. And he said, well, look at the automation in the automotive industry, or, you know, some of the manufacturing in... anywhere honestly, if... if you think about it, we've been doing that for so long, and we do such a great job. He said, you know, I can stand up a factory to build cars in no time, you know, maybe that's six months, maybe that's 12 months, I don't know what the actual timeline is. But, you know, if we try to translate that into life sciences, that's not the case. Part of it, I think, has to do with the fact that there is a lot of customization that takes place due to the complexity of the science, or the processes that we try to... to take on. But he was alluding to something totally different. And I think, you know, the key factor there is lack of standardization. We've done a much better job in that space over the last 20 years that I've been in this... in this career. You know, the SBS format for the labware that's being used in the labs, as well as the Silla standard for the instrumentation, has definitely helped, but I still think that we are really far behind in kind of getting everybody onto the same playing field for one reason or... or the other. So yeah, I... I think robotics is really still immature and there's a lot of room for growth in the life sciences space. 

Hannah Rosen: That's interesting, I mean, do you think that maybe some of the lack of standardization issues in life sciences comes from there being so many different providers that people are kind of mix and matching where they like. You know, you... if you think about a car dealership, you know, Honda, they control everything from start to finish, probably, of, you know, what they're doing. So, probably it's a lot easier to standardize when you're controlling all the different stages of the manufacturing process, whereas the life sciences, you've got so many... it's so specialized, you know, and we're not always communicating between the different stages of manufacturing. 

Olga Seltser: Yeah, and I think there was a bit of evolution that took place over the last, again, I can only reference the last 20 years. But I think we've seen explosive growth in the automation field, as far as life sciences goes, probably over the last 10 to 12 years. You know, there was a certain number of players in the... in that field. And then as part of, you know, healthy competition and technology advances, we started seeing a lot more players joining... joining that field, and then you also have generations that are now coming into the field, and they're equipped a little bit differently. Their knowledge and comfort level in adoption of that technology is different, right? Just like, you know, we weren't used to the cell phones in the past and we had all kinds of questions. Well, what about... what about security? What about privacy? What about so many different factors and similar questions remain within the scientists mind. How do I trust that the instrument is doing what it says it did, versus what I'm telling it to do? Just like every scientist is doubting which instrument is actually going to do... is actually going to do the job. 

Hannah Rosen: I guess the big question then is, you know, how do we fix this issue of lack of standardization? Because, you know, I know from my experience where, you know, once people find things... one thing that they like or have learned to use, even if they don't necessarily like it, if they know how to do it there's a lot of resistance in change and try and learning a new method, because it does take a lot of time and energy to invest in learning a new method, and especially, you know, you think of the financial investment and all that. So, is there a way that we can standardize some of these things? 

Olga Seltser: Yeah, I... I think you're absolutely right about the fact that people kind of stick with what they know. If something worked for them in the past, the next job that they go to, they're going to try to bring in the same equipment with them, right, because they know that that's a proven... it has a proven track record. But I feel that, that is typically the case for individuals that are science driven and their primary focus of job responsibilities is that, and automation is an enabling factor. I think where we can do a better job, and I think we do, do that today, is if they're automation folks present within those organizations, because they tend to have broader exposure. We attend conferences that are targeting a different set of folks, so we get to chat with the vendors and kind of, you know, better understand what might work, what might not work. It still comes down to first-hand experience in the lab, but I see like, as automation folks, we just have a better exposure to everything that's out there and, you know, of course, there's a little bit of personality component as well, and you'll have to make sure that you'll have an open mind, and you're always on top of any new technology that comes out and, you know, some people are a lot better at being the first adopters, whereas others will kind of sit back and say “let me watch this for a little bit. I might not have the time, I might not have the resources in the lab to take on something that just came out.” You know, back to your example with the car industry, it's like being the first person to buy, you know, the first off the release, right? Going back to your question about trying to standardize it... it's a difficult... it's a... it's a challenging ask, I think, because the science keeps on changing, and we try to have technology evolve to keep up with it. If technology would have been ahead, maybe automation could have been the one driving the standards.  

Because we know, for instance, that a strip tube is not an ideal type of labware to be used on automation, and there's, you know, so many different examples, like 1.5 mil Eppendorf tubes. Of course, we've become creative where we've customized the different racks, so we can still use them, non-automation friendly labware on our devices, to still enable some version of robotic handling, but we are making those changes to accommodate what scientists are used to. One thing that could be done better, and in some labs that is the case, having the scientists work with automation engineers from the get go so that the assay that's developed is never done in silo without sort of the considerations of, well, is that going to be amenable to the automation platform, right? So, having better conversation and having that collaborative experience, I think, would make it for a better transition to the automated platform. But again. I think standardizing is... it involves all of the laboratories kind of being on the same page, right, and driving towards a common goal. It also requires the vendors to be better aligned and how they achieve any... any final result. You know, there's... there's obviously friendly competition. So, we need to maybe get a little bit past that and understand that vendor A is going to be better in whatever they're better at, and vendor B is going to take off in a completely different direction and be better in... in that regard. But there's still some commonalities between how they get to that final result, if... if that makes sense. 

Hannah Rosen: Yeah, definitely. It's, you know, almost like, especially now as things become more and more specialized, one company is probably not going to be the provider of everything you need along the chain of laboratory automation, so it's all about finding a way so that vendor A is creating a product you can then, you know, maybe produce the data that then vendor B is creating a product that can then interpret that data, and the data, you know, they can talk to each other using a common language, so to speak. 

Olga Seltser: Exactly, yes, that's exactly the words that I was thinking about. The common language, right, cause it's great to have experts in their own area, but as long as what they hand off to the next step of the process is on the same language, that would make huge, huge impact in the space of laboratory automation. So, a little bit of that is seen when we try to integrate different devices with the robot arm, as being the common factor, right, that instrument has to be able to talk to all of the other third-party devices that are surrounding that robotic arm, and part of the delay in standing up laboratory automation is sometimes writing those drivers for an instrument that's never been integrated in the past. Or, you know, you can... you can think of a million other reasons why certain instruments that serve a certain purpose in the... in the world of science that are just not automation compatible because the product managers never thought of that instrument being integrated into a robotic system in the lab. So it's, you know, it's a... it's a paradigm shift on the vendor side as well to kind of think ahead and... and understand that there's a lot of utility for that instrument to be employed in a more integrated, higher throughput, more robust, you know, walk away type of environment. And maybe they don't get there for X number of years, but at least there's a road map, and they are working towards that end goal. 

Hannah Rosen: I would imagine too, you know, it has to be very intimidating for either like, you know, maybe smaller labs or academic institutions that may not have as much funding who, maybe, would like to move more into laboratory... the laboratory automation space and update their labs, but just... it's, you know, the cost is so huge and if you... you... it... it doesn't seem like you can start just, like, piece meal... well, I'll just buy one like, you know, I'll buy this robotic arm, but then everything else is manual or, you know, I'll buy this one liquid handler, but everything else is manual, and then every time you add a new piece of automation in there, it's almost like you have to completely start your processes over and re... rework them to fit into this automation process. So, you know, do you have any advice, having worked in, you know, all these different sized labs for these smaller labs who just may not have the budget to automate everything at once? You know, how... how can they enter this automation space? 

Olga Seltser: That's a great question, and it actually touches on a couple of different points, one of them being, automation does not need to be expensive, it does not need to be complex. I like to keep it simple, so maybe look at your workflow and try to identify either the most time consuming process or a step that introduces a lot of variability. You know, kind of try to identify what is your weakest link in that process, right, and maybe attack it from that angle, and there's plenty of solutions out there. So yes, there... it's a bit... it's a little bit of a challenge to investigate and try to figure out what would accomplish that task. But, I think kind of narrowing down the problem, and not... especially folks in that lab don't have much automation experience, don't try to automate from beginning to end, just find, you know, one step that you would like to improve and build on. From there, a little bit of your question also goes back to the educational component, and, you know, if some version of laboratory automation was a subject that was taught to the scientists, you know, early on, at least they would have been better equipped to understand what's possible, what's not, and what are the tools out there that could make their job a lot easier? 

Hannah Rosen: Yes, and it almost sounds like anybody who is studying and interested in eventually entering the, you know, biotech industry maybe would benefit from taking a biomedical engineering class or something of the sort. 

Olga Seltser: Yeah I... I think, and even biomedical engineering, it is so diverse and you can go into so many different concentrations, I honestly think it's a robotics class, you know, that kind of gives people the introduction to a very basic skill set, liquid handling, here's what you can do in the lab. And then, showing people the different flavors, between something that's simple and inexpensive, you know, a little bit more advanced. And then, sort of your final stage, when you have fully integrated systems, the nice evolution that we've seen with automation in general is the ease of use. So, if in the past you needed to really understand how to code and learn a particular, you know, proprietary language that these vendors were developing to drive their own robots. We've walked away from that, where it's... everything has become drag and drop, so anybody can be in front of the instrument and at least be able to get some basic steps completed. You might not be able to do, you know, more intricate work that these devices are able to... to handle, but at least that threshold or that activation energy that was required a long time ago is no longer there. So, I think we've made tremendous improvements in making laboratory automation more affordable and more accessible. 

Hannah Rosen: Definitely, and, you know, I wonder, especially with these smaller labs, and probably also as the technology gets easier to use, do we run the risk of, you know, falling into this trap of tribal knowledge? Of, you know, not having written out SOP's and, you know, having... once you lose the one individual who was running all this stuff in the lab, losing all of that knowledge. 

Olga Seltser: Tibal knowledge is an interesting concept that is always, I would say, deeply ingrained in laboratory automation in the life sciences. How that came about, I have no idea, and I honestly have never thought of it that way until, again, it came up in a conversation at... at that same dinner I was referencing. And we had, you know, we had some industry leaders in the space of automation at that dinner, and when that conversation came up, you know, we kind of looked at each other and said well, OK, that's job security. But at the same time, how much of tribal knowledge is potentially holding everybody back? Because as you said, you know, I've had the luxury of doing this for so long, and I carry that knowledge with me everywhere where I go. I try to do a good job at, you know, growing my teams, mentoring them, and expanding their skill set, and I love sharing that knowledge. But I can... I can imagine that not necessarily everybody has the same mentality. And, you know, there is competition involved. You know, who is the first to the market, so if automation enables the science, then how much of that tribal knowledge also plays into the overall success of the organization? So, I... I think that's definitely... it's... it exists. I would love for it to change so that, you know, every company is enabled from the get go to deliver good science and good medicines to the world. I just don't know if we're quite ready for it, to be honest with you. 

Hannah Rosen: Yeah, no, that’s... that's fair, yeah, and an astute observation, I think. So yeah, kind of looking towards the future, what do you see as being some of the risks to life science research if laboratory automation continues along its current trajectory? 

Olga Seltser: I think we need to do a significantly better job with technology, at least staying on pace with the science, or I would actually say to outpace the science to enable our scientists in the lab to do the amazing development and discovery that they set out to do as their career. I think we need to do again a better job at educating our workforce from the get go, right, and all of that together combined should enable innovation. Because I... I feel, you know, just recently, it might have been this week, that I was listening to Laboratory 101 Podcast, and the angle there was that automation actually enables innovation, and that's a very different angle to look at. I think many folks in the lab might disagree with that because again, they still question the utility, and the trust level is just not... not quite there. So, I think we need to do better on adoption of technology in the lab. It's here to stay, so I think we would serve ourselves a lot of good if we actually saw robotics as an enabler, as opposed to, you know, replacing the people that are doing the job in the lab. 

Hannah Rosen: Yeah, definitely, and it's... it is so funny because yeah, you hear that as a concern, a lot of, you know, am I going to lose my job as automation takes over? And... and I just always wonder how much are you really enjoying manually pipetting that you're afraid of losing that part of your job? 

Olga Seltser: Yeah, I... I think it goes back to the comfort level that's what people were taught in school, they feel confident that, that skill set is there to stay. And yeah, robots pose a threat, right? So, I... I think it's helping those folks to understand that it's going to require a different set of skills, but if you're open to that, then once you pick that up it's actually really, really difficult to find skilled automation engineers as it stands today, and I think part of it comes from the fact that, again, it's not offered as a major in school. It was less prevalent in the workforce, but that's changing. 

Hannah Rosen: I mean, it is interesting too though, because I feel like, you know, a lot of people get into science because they love learning and, you know, love learning new things and exploring new areas. So, I guess in some ways it is a little bit surprising that there's so much resistance to automation techniques because it's something new to learn, and it really broadens your ability to explore new areas of science. 

Olga Seltser: It sure does, yeah. But I think it does go back to that trust and, you know, there's a saying, if you want to do a job, well, do it yourself. So, maybe there's a little bit of that going on. For many people, robots are a black box. And, you know, you fall into two camps, either you don't mind that there's a black box that does this magic behind the scenes and you're sort of on the receiving end of the outcome, or you want to know every single little thing that goes on behind the scenes in the black box and unless you have control and full understanding of what happens behind the scenes, you don't trust it. 

Hannah Rosen: Yeah, definitely. It is tricky because we are getting to a point with technology where I know probably no matter who you are, you're not going to be able to fully understand every aspect of the technology because it just would take too much time, like, nobody has that much time to learn every aspect of this new automation technology. 

Olga Seltser: I was actually going to say that, you know, how much of that do you need to understand for you to be successful at your job? I guess it depends on what specifically your job responsibilities are, right? But maybe not having a full grasp of that magic happening behind the scenes, it's OK, you just need to be... you need to understand it enough to be comfortable for it, you know, to live side by side with you in the lab. And then, you know, hopefully as more companies understand the value that automation brings to the table, more of the automation engineers are hired or, you know, more scientists transition into that field, and it kind of makes up for the gap that there is today, yeah? 

Hannah Rosen: Well, we are almost towards the end of our time, so I just wanted to... to check with you Olga, and make sure, is there anything that we didn't get a chance to talk about today about this, you know, future of laboratory automation that you really want to cover? 

Olga Seltser: Yeah, I... I think, just want to go back to a couple of different themes that we had. You know, one of them is common language. You know, that would be a great change to see in the coming years between the different vendors to kind of get onto the same page so that the scientists in the lab using that technology didn't have to relearn a new skill set, or a new language, or a new software with every instrument that comes onboard. You know, that's definitely one of the topics to keep in mind. I think the other one which is huge for me is doing a better job on the digital component that goes along with automation. And by that I mean, you know, sample tracking and data acquisition and data mining. So, everything that LIMS, the laboratory information management system, brings on board, and data analysis tools. I think, you know, we need to do a better job in developing both anytime we bring one or the other, you know, it being digital or automation into the lab, they go hand in hand, they need to live together. We need to understand the implications of cutting ourselves short in either one of those directions. 

Hannah Rosen: Yeah, lots... lots to think about, that's for sure. 

Olga Seltser: Absolutely, but I think that's what makes this field so exciting. There's so much potential, there's so much growth and opportunity, and being in it we can influence where it goes. And I think that's part of the rewarding experience is making those changes and enabling the technologies further down the road. 

Hannah Rosen: Definitely. Well Olga, thank you so much for taking the time to talk with us about this really important topic. I really, really appreciate it. It's certainly, like I said, a lot to think about, and I really hope that we'll see you at some of our future SLAS events so that we can keep this conversation going. 

Olga Seltser: Absolutely. The pleasure is all mine. Looking forward to SLAS2023 that's in San Diego coming up very shortly, but I really do appreciate the opportunity to have this conversation. And you're right, it's a very important one.