BioBuilder Career Conversation: Prasanna Neti Transcript

The way these Thursday sessions will work is that we have the nice opportunity to work and speak with somebody who’s in the field of biotechnology, biomanufacturing, synthetic biology, bioengineering, engineering, writ, large science, engineering and, with those conversations, just open up, pull back the curtain a little bit on what these kinds of careers look like and how people get there.

Okay, Awesome, All right. So my name i PresRato Neti. I work for. I’m a principal automation engineer at Stellix Global Services. I’ve been working in the industrial process automation industry for over 19 years and mostly in life sciences, pharma, and I’ve worked in R&D, clinical and manufacturing settings. I’ve also done some work with specialty chemical and pulp and paper, but that’s kind of not part of this. I have a bachelor’s of science in chemical engineering from Georgia Tech and I got a master’s in business administration from Babson 15 years later. So just to let people know, it’s always great to go back to school and expand your learning as you develop in your career. My hobbies are gaming. I like to puzzle I’ve been on a puzzling kick lately Music, and I’m a classic movie fanatic.

So, that’s sort of just the about me. In terms of my career. I started out after college working for Honeywell Process Solutions. They have a huge industrial process automation business, so it’s kind of like a small fish, big pond. I started out as an individual contributor and moved into leading smaller projects where we do deployments for customers of putting these automation systems in so that we could optimize processes at sites. Now I work for NECI, which is now Stellix Global Services we just rebranded today.

So very confusing. If you look me up on LinkedIn you might be a little confused, but so I listed both there. I started as a senior systems engineer, did that for about seven years and then wanted to kind of get more into the commercial side of things. So I moved into project management. I kind of grew through the project management organization for about four and a half years and then I decided, hey, I see technology really pivoting with Gen AI and all of this stuff kind of coming about. This is really going to impact how we work in lab space and we work in manufacturing and all of these different spaces. So at that point in time I was like ready to make the move back to being back to engineering, and so that’s how I ended up back as a principal, as an automation engineer again.

Yeah, and that’s kind of why I was excited to present, because when I started in this field I didn’t really know how big it was, how much there is to it. So just a high level overview. When we’re talking about process automation on the industrial side, we’re talking about replacing manual steps with automated workflows, placing manual steps with automated workflows. And we do this through hardware, so you could think of transmitters in the field, electrical components for those transmitters for you to get the signal to a actual screen that you can see, the servers, the networking. There’s a whole digital stack that needs to be built for all of this data and information to come through, to be built for all of this data and information to come through.

It’s definitely something that operators interact with and that’s something we definitely take into account when we’re designing systems. Another big reason why you want to automate is data. Everybody wants to have data on their runs, right? If you do things manually kind of are constricted to your Excel spreadsheet. You might be typing things into and manually managing, but with automation all of those values and set points and process parameters can all be brought into a central database and tracked and historized and you can kind of see over each run how you’re doing and optimize your process accordingly. And I think I already talked about scale up. So there is something to be said about safety as well. When you’re talking about scaling up a process, you’re working with big equipment like huge pumps, huge valves, and it’s using automation can make that safer because you can build alerts and alarms to notify you when your process conditions are starting to get unsafe and you can start, you know, making work plans around that. Hopefully that doesn’t happen.

And that’s why you put automation in, but it’s another reason why automation helps. So I can go into this slide, but maybe I will. I’ll just touch on this a little bit. If I’m going over anything, just let me know, by the way.

Okay, so, in terms of, I talked about a digital stack before and it wasn’t very clear when I mentioned what that was, but there’s a lot of systems that you can use to optimize and build your process out. So when you have a process and you know the steps you’re going to have to put, have transmitters that read the different values coming from the various process parameters, you’re monitoring over the course of your process. An example of this is on a bioreactor. You want to monitor pH on a bioreactor to make sure that your cells aren’t, you know, living in an environment that is not happy for them. You want them to keep producing and in order to do that, you monitor the pH and if that pH goes out of line, you have equipment that then responds to that.

So, on the field level, we’re talking about, like all of the devices, actuators, sensors, valves. They’re valves that can open and adjust and close based off of different set points you want things to be at. So we’ll go a little bit more into that in detail. My background is kind of at this mid-level here between the control level and the supervisory level. So I’m good at taking those signals, putting them into a central place and then building machine interfaces for people to work with so that they can interact with their processes through digital means. So on the control level you have a lot of computing powers where you have your controllers, which are essentially processors that are not on a Windows OS because Windows crashes and you don’t want your plant to go down.

So you, you put, you put all of this stuff, you have all your computations, your calculations, your signals coming through a very, very centralized controller level type layer. Then you move all of that up to what I call the HMI layer, which is human machine interface, and that’s where people can interact with this information that they’re getting from the field. That information is important for operators to see, to see if their process is within the bounds of where they want it to be. It has indicators like alarms, notifications for when things are getting out of range and out of spec. So that’s really kind of where my background is and where I’ve lived for a long time.

There’s so much data. You can make so much data just from one batch run, but you have to retain that data because if something down the line happens to that batch, if there’s a recall on the product or anything like that, you need all of that historized data in order to go back and reference and see, okay, well, what were the missteps here and take it from there. So it’s a compliance thing, but it’s also data can be used to optimize your process. It can be used to especially now with generative AI, people are starting to build models to basically take okay, well, this is how my last run went. This is how much I used of everything all my raw materials. What is the best? Here’s data over 10 runs, how, what are the best conditions? What should we be doing to make this to get the most out of our raw material?

The next two levels that are higher up are my company also works with, but are not really my area of expertise. We have a manufacturing execution system. This is a operator interface that operators work with to assign batch IDs, assign codes to their runs so that when you and a batch ID is essentially like a number associated with your process run, when you’re making your drug or whatever you’re making, so that you can kind of backtrack it over time, like you can say, oh, this batch like, and again, it’s all, for you know when you distribute this stuff. If there’s recalls or anything like that, you have to be able to know, hey, which batches were impacted, and that particular ID gives you that information. You can also have all your historized data under a certain batch ID, which makes it a lot easier to search and find. Mes systems also help you track your raw materials going into your process. You build a bill of materials when you in your MES, when you start your process run, you start over the course of your run, you you start entering in data, or sometimes it automatically populates if you’re building your, if that’s how you build your stack, with what raw materials you used and how much product you made at the end of the day.

So MES systems do a lot of different things at the operational layer, but to kind of just boil it down to the two basic things that I’m mostly familiar with, that’s where I think they add a lot of value.

And in terms of enterprise resource planning, that’s an ERP, that’s a business system. So all systems have like a centralized business system they use to monitor costs, to monitor how much money they’re making and all of that. So what an ERP can do is extract data from an MES and see okay, well, how much is it costing me, or in a batch, and how much am I making off a batch, and you can kind of optimize your costs that way too. So there’s a whole stack of things systems, I guess and each one of these is a system in and of itself that you have to integrate in order to get that full stack experience. So that’s kind of in a nutshell. I kind of deep dive into my area of expertise, because it’s this is where I sort of what I know but and I kind of talked about this before with the transmitters.

So this is that first field level that I was talking about. You have flow control valves, level sensors, steam valve, here they’re all going into this control level which is your, which I have here what is called a PLC, that’s a programmable logic controller, and it’s one of those types of processors I was talking about. It doesn’t look like a server, it doesn’t look like a PC, it’s a. But it’s a computer and it’s sitting in a cabinet in the field and it’s doing its calculations and pulling the data and information so that it and then pushing it out to the, to what we call a SCADA system, and SCADA stands for Supervisory, Control and Data Acquisition. And it does just that. You can control your, you can, you can see the data, you can control your system from the system, you can acquire data from it, you can build trends, you can build charts to see how your runs went.

One other thing I wanted to kind of point out, because one of the kind of basic functions of controls engineering is control. How do you control something? So we have this terminology called PID, which is proportional, integral and derivative. It’s a basis from which you do control, and the best way I can explain it without getting too technical and too mathy is a thermostat. So when you have a thermostat, you set it at a temperature and if your room is below that temperature your heater kicks in. But then what causes that heater to stop? Well, you set a set point and once it sets that set point, there’s logic in the thermostat that says, hey, I’ve reached my set point, I don’t need to be on anymore, and it turns off, and then, once your temperature goes down again, it comes back on, and so that’s sort of how. That’s sort of like a very simplified version of how process control works. But you can have this type of control on a device level At this control level. You can have it in different places, but generally you’re controlling your process in this way, and this makes it easy for operators to put a set point in for a particular process parameter and not have to keep monitoring it and adjusting their flow or their pressure or their temperature manually to accommodate the process, to keep things at the right set point.

So, and last but not least, this is my last slide I promise I wanted to bring up a bioreactor. I’ve worked with so many bioreactors. My company has a bioreactor branch and so for me, bioreactors have a very. They have a special place in my heart. This is an example of kind of an industrial scale bioreactor. I think in one of my other pictures I showed more of a bench top bioreactor, but this is. These are the.

This is the type of equipment I’ve worked with in the past and they can be up to like 1,000 liters, 5,000 liters, they can get big and as you’re growing your cells you move them to different reactors so that those reactors can accommodate the size of the growth that you’re wanting to hit for your batch. So some of the equipment here that I’m talking about are on the field level of that pyramid that I showed before. We have a feed pump that puts, if you’re inoculating your batch or you’re putting media into your vessel to start your batch or you’re putting buffer in to control pH, you would do that through like a feed addition. So you put that into the top of your tank and then you have processed gases which you monitor, and you also have different set points for it, depending on your process. You might use air, you might use oxygen, carbon dioxide lots of different variables there that you can tweak, and it’s very process contingent.

So I just listed all of the things you would use. Not every process uses that. And then I noted the agitator, which is essentially like a stirring blender that is on a but with a huge motor that is on a big vessel, and what you do with that is you control speed.

So you give it a speed set point and it runs at that particular speed, or you tell it hey, I only want you to run at these certain points in the process and, based off of coding, it knows when to run and when to stop. Then we have the regular sensors, like vessel level pH conductivity. Those are just values coming into the system that we monitor and then also can control with through process gases, through additions and things like that. So if your pH is going high, you might want to add some acid, right?

I’m trying to like, think you want to bring it down, to make it more acidic.

It didn’t happen. So today’s talk is all about laboratory notebook keeping and process and data collection. So thank you for teeing that up. The other thing that you’ve set up very nicely for us is that on Saturday, the students are going to come back into the lab and they’re going to start to grow some cells. They’re going to use it on solid media, on liquid media. Now, of course, we’re doing it on a very small scale, but I love that with this presentation, you’ve started to introduce what industrial scale manufacturing of biological materials can look like. So the connection between what you do as an industrial process engineer and what the students are learning today and Saturday as we scale up the cells to make things that we want them to make, is really incredibly valuable. So thank you, thank you for taking the time to talk with us. You are more than welcome to stick around. We are going to cover some things Now.

Maybe, if there is a question, one question. We have time for one question from a student. If you wanted to maybe stop sharing your slides, and if a student wants to either come off mute or join us, you know, or put something into the chat, let me see you know, or put something into the chat. Let me see, is there anyone? Is there a student who would like to ask a question in the chat or by coming off of mute? They are probably a little shy. It might be the first day today. Yes, it is the first day today Online, rather. Yes, yes, thank you, christopher.

Well, thank you for a wonderful talk. It’s really fascinating what you do and it is an interesting pathway for somebody who’s super interested in science. Right, it is. You know the words operation and processing and things like that. It’s a great combination of engineering and science. So, yeah, thank you. Thank you so much for being here.