Pod Insights Inspire AI: Transforming RVA Through Technology and Automation Ep 1 - Pilot Episode with Will Willis from AI Ready RVA and Jason McGinthy from Capital One About this episode Send us Fan Mail Want to join a community of AI learners and enthusiasts? AI Ready RVA is leading the conversation and is rapidly rising as a hub for AI in the Richmond Region. Become a member and support our AI literacy initiatives.
Transcript Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:03 Jason McGuinthe? Today's episode is made possible by Modern Ancients, driving innovation with purpose. Modern Ancients uses AI and strategic insight to help businesses create lasting positive change with their unique journey consulting practice. Find out more about how your business can grow at modernancients.com.
0:25 Speaker 01 Hey Richmond, welcome to Inspire AI, where we spotlight companies and individuals in the region who are pioneering the development and the use of AI. I'm Will Willis from AI Ready RVA, where our mission is to cultivate AI literacy in the greater Richmond region through awareness, community engagement, education, and advocacy. Today, we have a special guest, Jason McGinthy from Capital One. Welcome to the show. Could you please take a moment and introduce yourself and tell us a little bit about yourself and also Capital One?
0:55 Jason McGuinthe? Absolutely. Thanks for having me on the show, Will. I'm a Senior Manager of Software Engineering at Capital One, worked for the company for about 13 years. As a Senior Manager, my teams integrate machine learning models into the experiences that we build within the platform that we manage. The, my responsibilities include designing, developing, and integrating with the machine learning models, as well as ensuring that the engineering teams are thriving with their growth mindset, and they have the access to the tools that they need to accomplish their
1:38 Jason McGuinthe? job. And I also ensure the scalability and efficiency of our systems so that we can maintain our competitive edge.
1:49 Speaker 01 Thanks, Jason. What inspired you to start working in the AI field and what drives your passion for this technology?
1:56 Jason McGuinthe? Yeah, that's a great question. I think back to my undergrad years in a class, I think it was called Decision Management Systems. I was doing a report on artificial intelligence, and I was awestruck by its potential. And this was many, many years ago, so well before generative AI. I felt like...
2:22 Jason McGuinthe? Being able to automate systems and decisions was gonna change the world eventually. So I tracked it over the years to some extent, not really getting involved in it because I didn't have the means to make the impact in my job. Over the years, though, I did transition through roles that got me closer and closer to artificial intelligence to where I am today. I had to take many lateral steps, a few vertical leaps as well. And here I am, working in artificial intelligence and hoping to make a big impact to the customers in the world around us with it.
3:04 Speaker 01 All right. So tell me, what type of AI technologies or solutions do you enjoy working with? And what problems are they designed to solve?
3:12 Jason McGuinthe? Well, in most recently, I've been working with natural language processing.
3:20 Jason McGuinthe? The problems that I enjoy learning about and solving are really in the the form of conversational AI, business-related problems. So it's got to have a purpose. When when I think about it, like, the problem that we're we're trying to solve is giving customers access to information
3:44 Jason McGuinthe? in an automated way so that we're not spending as much on on resources, human resources. That's essentially what the problem is, that natural language helps pretty much all industries. What's interesting about it to me is the capabilities of of speech to text and text to speech, sentiment analysis, being able to leverage content that formerly formerly was more challenging to deterministically feed a chatbot. And now it's very streamlined. And finally, I think that document summarization has a lot of runway for customer-facing
4:29 Jason McGuinthe? solutions, and I'd like to see that develop in in companies more.
4:36 Speaker 01 And what type of AI technologies or solutions do you enjoy working with? And what what problems are they designed to solve?
4:44 Jason McGuinthe? Sure, yeah, I'm a big advocate of leveraging natural language processing as a technology to solve many business challenges.
4:56 Jason McGuinthe? Interpreting customers' needs, leveraging the sentiment to make realtime decisions. I think those are really interesting problems to solve. And I like studying the frameworks around, especially machine translation. Because I think the world could be a much more unified space if everybody could communicate well across various channels. And I think machine translation is going to change that for everyone.
5:37 Speaker 01 Awesome. Now, let me ask you this, Jason. How do you see the importance of data privacy in AI implementation?
5:45 Jason McGuinthe? Yeah, that's an important question, for sure. My instinct is to start with building trust with your customers. And that is cornerstone to data management. You must build an environment that customers' data is safe and secure, within their control. They should be able to opt in, out of policies and understand those things. Clear language, so that they know what they're getting into when you're offering them solutions.
6:19 Jason McGuinthe? I also think that companies should take a step further in building a well-managed data ecosystem that ensures these things are governed properly and the data is easily accessible and usable by all parties. Because if your data is not available in that manner, then how can you expect a machine to leverage it properly for you?
6:48 Speaker 01 Alright, so let's talk about what are some of the biggest technical challenges teams may encounter while developing AI solutions? And how would you recommend them addressing?
6:58 Jason McGuinthe? Alright, that's a, that's a good one. So you want your solutions to have high availability, scalability. You want to choose the right algorithm to address the
7:16 Jason McGuinthe? the problem you're solving. You want to ensure that the system is fed with quality data.
7:30 Jason McGuinthe? And that your training
7:34 Jason McGuinthe? and inference optimizations are well suited for your, the environment that you're operating it in. So I think some of the, some of those challenges can be overcome by hiring the experts that can do it for you. You have your researchers, your data scientists, your engineers. You also should consider really full solutions. Anything in the cloud typically scales well if you build it properly. Selecting the right algorithm, I think, is
8:10 Jason McGuinthe? a challenge in and of itself because there's so many out there.
8:16 Jason McGuinthe? You might want to consider using pre-trained models that that standardize things and then fine tune according to whatever solution you're trying to solve.
8:30 Jason McGuinthe? And I guess finally, just embed continuous learning into everything that you're doing so that you are never satisfied with the status quo and always looking to improve your efficiencies, which is performance and accuracy of your models as you go.
8:48 Speaker 01 That's a great segue into my next question. What comes to mind when evaluating performances of a model?
8:56 Jason McGuinthe? Top of mind, accuracy, precision, recall. All those have various benefits. When evaluating models, I would say. They all serve a purpose and depending on what problem you're solving, you will might want to wait one more than the other. Typically, more than one is a good idea to pay attention to in the first place. The other thing I would say outside of, you know, model performance is system performance, so your scalability and. During runtime, you want to make sure that you're. Leverage the right amount of data to solve problem.
9:39 Jason McGuinthe? And you want to have the right computational resources. To deliver the inferences. According to.
9:52 Jason McGuinthe? Performance expectations.
9:55 Jason McGuinthe? Yep.
9:56 Speaker 01 All right. My next question, what trends do you foresee in AI development?
10:01 Jason McGuinthe? Top of mind. A few. Modal. Being able to leverage AI technologies across various. Channels and and solutions, meeting the customer where they're at.
10:17 Jason McGuinthe? Whether it's speech or typing. Image generation. All can play a role in meeting a customer's need. Then you have edge. Based AI solutions where. Models are getting closer and closer to the devices and being even built into the operating systems.
10:43 Jason McGuinthe? That's going to play a big role in the future of AI. Terms of reducing latency and cost.
10:51 Jason McGuinthe? And then I've always liked reinforcement learning because it takes me back to like
10:59 Jason McGuinthe? a kid when I think about robotics and and being able to to train a robot. How do you do that? Well, you encourage it to do more of what you want it to do and you you discourage it from doing things that you don't want it to do. And that's how you reinforce its learning behaviors. And I think that's really cool stuff. Um, haven't really gotten my hands on that yet, but uh, definitely fascinating. And then finally, AI agents, you know, autonomously being used in broader uh, and more complex problems. I think that being able to leverage agents to think or reason and make decisions on behalf
11:43 Jason McGuinthe? of customers and just validating, allowing them to validate that the customer wants the agent to execute it on their behalf. That that I think we're going to see a lot of in the in the coming years.
11:54 Speaker 01 Right. What advice would you give an aspiring AI developer or companies looking to enter the AI space?
12:02 Jason McGuinthe? You want to start with understanding the data because your AI solution will only be as good as the data you feed it.
12:10 Jason McGuinthe? You want to follow industry experts, find, find people that
12:17 Jason McGuinthe? have similar values in business and personal life that are leveraging the the tools and technologies to learn from them.
12:28 Jason McGuinthe? Read their blogs, their books, listen to their talks, all of that good stuff. You want to stay up to date with communities that are locally available to you like AI ready RVA, right? Um, and that's how that's how I'm building my outside of Capital One network today. Um, and finally, definite definitely understand the ethical implications. You want to ensure that you recognize the bias in your systems, uh, any privacy concerns and uh, know that these things aren't always uh, transparent and to really
13:12 Jason McGuinthe? understand how your systems are behaving and and evolving as you go.
13:17 Speaker 01 All right, one last question before we end. Um, not AI related, but if you had any superpower, what would it be and why?
13:27 Jason McGuinthe? Hmm, yeah, that's a good one. So my superpower would have to be like Doctor Strange's superpower, able to travel through space and time through various dimensions. I think that would be a lot of fun. And, you know, I'm a, I'm a big fan of the mystic arts. So, yeah, let's go, Doctor Strange.
13:56 Speaker 01 Awesome, awesome. Well, thank you so much, Jason, for being on our show today and sharing your insights. Thanks to our listeners for tuning in today. And if you or your company would like to be featured on an Inspire AI Richmond episode, please drop us a message. Don't forget to like, share and follow our content to stay updated on the latest news and events from AI Radio RBA and see you next time.
Next Ep 2 - AI's Impact on Work and Life: A Conversation with Jack O'Gorman Page updated October 26, 2026 · transcript and summary are machine-generated