Q&A with Basetwo AI’s CEO Thouheed Gaffoor

Thouheed Gaffoor. CEO at Basetwo AI.

Canadian Manufacturing spoke with Thouheed Gaffoor, founder, investor and CEO at Basetwo AI,  following the company’s introduction of its no-code manufacturing platform for the pharmaceutical industry.

Basetwo’s no-code platform helps data scientists and engineers work together without need for programming language like Python. Instead, users of the platform can create digital twins of their plants using the drag and drop feature on the interface. 

Digital twins are not new to the manufacturing world; however, Gaffoor explained how Basetwo anticipates this platform will help streamline the manufacture of pharmaceutical products.

Q: Where and how has this technology been deployed so far?

A: We’re still actively piloting. I’m under NDA, so I can’t mention which customers are using it, but it’s intended for pharmaceutical manufacturers and chemical manufacturers.

Q: Was there a lot of pressure from the industry to create something like this?

A: Yes, so let’s say you’re talking about a monoclonal antibody. For a pharmaceutical company, there are a lot of different process steps in the plant itself.

The whole process of manufacturing is not like one or two steps. It’s actually dozens of discrete pieces of equipment that are taking some upstream material, doing something to it and then producing an intermediary product that then goes to the next step.

You’ve got to control for the quality coming out of each intermediary step and make sure that that step is actually operating as efficiently as possible. If it’s not, you’re going to be wasting a lot of material resources. You risk being non-compliant with quality when you measure the quality at that intermediary step.

Q: What what kind of impact do you anticipate this will have on pharmaceutical manufacturing?

A: There’s a couple of things. When you’re data driven, you can  improve the reliability or predictability of your product quality. If you think of where the manufacturers are today, maybe their product quality has wider distribution.

If you use our platform, you can tighten up that distribution so that your product quality is more predictable and is essentially better. It’s a narrower distribution, so there’s less variability in that quality.

The second piece is improved process understanding. If you’re collecting all this data and characterizing your manufacturing process, you’ll improve your understanding of how each individual manufacturing step will impact that final product quality, which is important from an FDA and regulatory perspective. You need to be able to demonstrate that understanding to the regulator.

The third piece is the efficiency metrics. With this type of system, you can increase your overall yield. That could be anything from you know, a 5 per cent increase in yield up to a 15 per cent or 20 per cent increase in yield. That can have a significant impact on profitability as a company.

Q: What sort of feedback have you received from partners in this industry?

A:  A lot of manufacturers understand the need for that digital twin concept, and have started building their own in-house AI technology. Some of the feedback we were getting was that their attempts were not scalable.

Maybe one research group in Europe was trying, but they were unable to then operationalize that and scale it across manufacturing sites in the U.S.

There was a need for a scalable technology that can operationalize those digital twins, across the entire organization, so that was a big aspect of how we thought through building this product.

In that same vein, we found that certain clumps of the organization were ahead in terms of their understanding of processes, technology, and R&D overall.

So, how can you make that knowledge more accessible across the entire organization? That’s something that we baked into the platform. That’s why collaboration is such a big piece of the Basetwo product because we’re actually getting that feedback about how can make this operational, scalable and make institutional knowledge more accessible across the entire organization.

Q: What do you want others in pharmaceutical manufacturing to know about this platform?

A: The key takeaway is that this isn’t really about replacing their in-house expertise or engineers or data science efforts. It’s really about augmenting and giving them a platform so that any efforts that are regionalized or local can be scaled across their entire organization. It’s about giving them a toolset or framework that they can use to operationalize AI and digital twins in their organization.