Powering Seamless Customer Experiences With Shawn Bowen

Shawn Bowen is the CTO of Quik!, a company specializing in enterprise forms processing solutions that empower financial advisors and businesses to digitize workflows and automate complex form submissions. He brings over two decades of experience building high-performance software systems and leading engineering teams. Before joining Quik!, Shawn held leadership roles, including VP of Software Development and CTO at other technology organizations, where he developed advanced data and analytics platforms. At Quik!, he focuses on improving reliability, scalability, and customer experience through API-driven architecture, automation, and AI innovation.

Here’s a glimpse of what you’ll learn:

• [2:11] Shawn Bowen discusses Quik!’s mission and how API-first approach enables customer success

• [3:54] Impact of outages on customers and importance of reliability

• [6:28] Turning failures into growth through feedback loops and post-mortems

• [9:29] Inside Quik! 911 incident response and team collaboration

• [11:05] CI/CD pipelines and contract testing to prevent breaking customer integrations

• [15:10] API complexity and challenges of maintaining backward compatibility

• [18:22] Shawn talks about the evolution of technology and AI as the next major revolution

• [28:12] Using AI to understand codebases and improve system reliability

• [33:07] Balancing AI capabilities with data privacy and responsible usage
  

In this episode…

Modern businesses rely heavily on complex software systems, yet even small disruptions can halt operations and frustrate customers. Maintaining reliability while continuously improving and scaling technology is a constant challenge. How can companies innovate rapidly without breaking the systems their customers depend on?

Shawn Bowen, a seasoned technology leader and software architect, explains that the key lies in building strong feedback loops and resilient processes. He emphasizes conducting root-cause analysis after failures, implementing CI/CD pipelines with safeguards like contract and playbook testing, and deeply understanding how customers actually use products. Shawn also highlights AI’s role as a force multiplier — helping teams improve code quality, accelerate development, and reduce errors — while stressing the importance of treating AI as a tool that still requires human oversight and context.

In this episode of The Customer Wins podcast, Richard Walker interviews Shawn Bowen, CTO of Quik!, about improving customer experience through reliable software and AI innovation. Shawn discusses incident response culture, API consistency, and proactive testing. He also explores AI-driven development, developer productivity, and balancing innovation with stability.

Resources Mentioned in this episode

• Quik!

• Richard Walker on LinkedIn

• Shawn Bowen on LinkedIn

• "[Emerging Tech] Eliminating Data Complexity for Growth With Mark Ovaska" on The Customer Wins

• "[Emerging Tech Series] Transforming Financial Advisors’ Workflows With AI Agents With Chirag Gandhi" on The Customer Wins

• "[Emerging Tech] Simplifying Operations in Wealth Tech With Max Klein" on The Customer Wins

Quotable Moments:

• “Mistakes only become true failures when they aren't acted upon, and you need to close the feedback loop.”

• “As an API company, our solution tends to only be visible when there is a problem.”

• “You just don't know how people are going to use your APIs in novel and unique ways.”

• “AI creates a really good tutorial for them on how to architect a system.”

• “Think of AI as a tool and assistant, maybe even an employee, but you're the manager.”

  
  

Action Steps:

1. Build strong feedback loops after failures: Capturing issues, analyzing root causes, and feeding insights back into development helps prevent repeated mistakes while ensuring continuous improvement and turning outages into long-term reliability gains.

2. Implement contract and playbook testing: Validating not just APIs but real customer usage patterns reduces the risk of breaking integrations and protects customer workflows while maintaining trust as systems evolve.

3. Create a rapid incident response culture: Mobilizing teams quickly during outages minimizes downtime and customer disruption while building accountability and reinforcing a customer-first mindset.

4. Leverage AI as a development assistant: Using AI to review code, guide architecture, and automate tasks increases speed and quality while allowing teams to focus more on solving meaningful problems rather than repetitive work.

5. Invest in context-rich AI systems: Training AI with relevant company knowledge improves accuracy and usefulness in real scenarios while enabling better decision-making, faster onboarding, and more effective customer support.

   
   

Sponsor for this episode...

This is brought to you by Quik!

At Quik!, we provide forms automation and management solutions for companies seeking to maximize their potential productivity.

Using our FormXtract API, you can submit your completed forms and get clean, context-rich data that is 99.9% accurate.

Our vision is to become the leading forms automation company by making paperwork the easiest part of every transaction.

Meanwhile, our mission is to help the top firms in the financial industry raise their bottom line by streamlining the customer experience with automated, convenient solutions.

Go to www.quickforms.com to learn more, or contact us with questions at support@quikforms.com.

Powered by Rise25 Podcast Production Company

Episode Transcript:

Intro: 00:02 

Welcome to The Customer Wins podcast, where business leaders discuss their secrets and techniques for helping their customers succeed and, in turn, grow their business.

Richard Walker: 00:16 

Hi, I'm Rich Walker, the host of The Customer Wins, where I talk to business leaders about how they help their customers win and how their focus on customer experience leads to growth. Some of our past guests have included Mark Ovaska of Precept. Chirag Gandhi of Mili and Max Klein of Lea. Sorry, Ellie. Ellie a.

It's an AI company called Lea. Today I'm speaking with Shawn Bowen, the CTO of my company. Quik!. And today's episode is brought to you by our company, Quik!, the leader in enterprise forms processing.

When your business relies upon processing forms, don't waste your team's valuable time manually reviewing the forms. Instead, get Quik!. Using Quik!. You'll be able to generate completed forms and get back clean, context-rich data that reduces manual reviews to only one out of 1000 submissions. Visit quickforms.com to get started.

 All right. Shawn Bowen is the chief technology officer here at Quik!. And he brings over two decades of experience transforming complex ideas and elegant high-performing software solutions. Before joining Quik!, Shawn has served as the VP of Software Development at Optimal Efficiency, where he led development of advanced digital artifact management systems for the US Air Force Research Labs, and he was previously CTO at Owl ESG, where he architected a pioneering analytics platform that transformed how environmental and sustainability data is used by institutional investors. Man, I'm happy to have you here today.

Shawn, welcome to The Customer Wins.

Shawn Bowen: 01:47 

Thanks, Rich, and thanks for having me.

Richard Walker: 01:50 

Yeah. So for my audience, if you haven't heard my podcast before, I love to talk with business leaders about what they're doing to help their customers win, how they built and deliver a great customer experience, and the challenges to growing their own company. We have amazing leaders here at Quik!, so I thought it'd be nice to bring Shawn here. So Shawn, let's gain your perspective on our business. How does our company help people?

Shawn Bowen: 02:11 

Yeah, I think that's a great question because at the end of the day, that's our mission, right? And it's important that we all have a coherent mission. And when I first came on to Quik!, it was a little unclear to me exactly what the value was that we were adding. I knew that we were creating digitization of forms, and we were enabling our clients to incorporate that into their own systems and workflows. But what really occurred to me is that as an API first company, we're actually enabling really powerful automated services that couldn't exist otherwise.

By and large, we tend to have the lion's share of all of the financial advisor forms within our library, and I'm not aware of any other company that has quite that breadth of scale. So we power quite a few of the technologies that financial advisors rely on, on a day-to-day basis to get their work done. And we hear about it. If there's ever any disruption in service, that means that tens of thousands of people can't actually conduct their business and do their job. So it's something that I take very seriously. Without us, a lot of the work that's getting done is either going to be incredibly inefficient or just won't actually happen at all.

Richard Walker: 03:34 

Yeah, it's been pretty crazy over the past 20 plus years running this company. Just how much of an impact we have. And, and unfortunately, you don't know it until you take it away from somebody and they're really, really upset. And, you know, I hate to admit it, but we've had those moments while you've been here. And how has that felt to you when we've gone through those moments.

Shawn Bowen: 03:54 

Oh, it feels terrible. I mean, when we're letting down clients. I feel like that is an obligation that we have. Now we've failed to deliver. But there's the flip side to that.

There's the time, the 99.9% of the time where we are doing a great job. And, you know, it's just like internet reviews. You're going to hear about all the bad stuff because that's what motivates people to reach out and activate. You don't often hear about all the good stuff. And so it doesn't become clear until you really start doing the metrics and you see that, you know, we're delivering on our promise at a very high rate.

And that's what makes me feel good and proud about the work that we do at Quik!, is making sure that our clients almost don't even remember us. I know that sounds weird, but as an API company, our solution tends to only be visible when there is a problem. When things are going great. That's the best solution for us. And people don't ever think about Quik!.

So in some ways it's you know, it's great to have her name out there, but it's also wonderful when you haven't heard from a client for a few months. And when you do contact them, they tell you everything's going great. So yeah, those times where things aren't going wrong, those are important for us to remediate and take seriously. But I try not to bring down the overall experience of the value that we provide it Quik!.

Richard Walker: 05:25 

Yeah. A mentor once said to me, Rich, buy a suit. Nobody's going to comment that's not what you want them to notice about you. Right? So it's kind of like that.

We don't want Quik! to be noticed all the time, because who cares about the brand of paper they're filling out on or working on? What we care about is did they get to do the work they need to do in the most efficient manner? Was it a good experience versus a terrible experience filling it out? Let me go back because look, on this show I typically talk about the customer wins and we tend to be very positive. And I didn't plan this, Shawn.

It's just coming to light. I'm thinking about this going, we never talk about the bad. Yeah. And so when we have a bad, the bad is an outage. A bad thing is a hiccup.

A customer can't do what they need to do for some period of time. I kind of want to kind of play this out. We know it's a bad experience for the customer, but what do we do internally? How do we solve that? How do we take that feedback and change and evolve and grow?

What do we do in that experience to make it a better both kind of navigate that difficulty and have a better outcome down the line?

Shawn Bowen: 06:28 

Yeah, I think that's a yeah, that is the crux of turning something bad into a learning and growth opportunity, right? There's, there's mistakes that will be made through the course of any career and any business enterprise. That's unavoidable. But the mistakes only become true failures when they aren't acted upon. And what I mean by that is that you need to close the feedback loop.

So we started on the journey by producing a product that our customers use. That product will invariably have issues with it, either bugs or user experience issues that we could improve upon. And what we want to do is when those circumstances arise and we're notified of them, or we're proactively aware of them, that we add that back into the feedback loop of the software development life cycle. So it's important to capture the failure and then have a post-mortem on the failure. Understand what was the cause of that failure.

The root cause is what we usually call that. And then we try to identify different ways in which we can mitigate our risk of having that root cause occur again. Sometimes it's pretty easy. Software bugs that are just obviously discernible. We can fix and we can just take care of those in a patch.

Others that tend to be more complex are things that are process related. Those are the ones that I don't want to say keep me up at night, but those are the ones where I really want to spend my time to ensure that from a process and a culture perspective, that we don't continue to make the same mistakes. And if we can do that successfully, if we can produce the product, evaluate the risk, capture any sort of issues that occur, determine somehow to mitigate that and feed that back into the lifecycle, then we've won. And I believe that when we win, the customer wins. So when you take a look at how we handle any, any of the bad is we try to turn it into something good.

We try to turn it into a learning opportunity, a growth opportunity. And from a software perspective, just making sure that our software is more safe, more scalable, more enterprise ready.

Richard Walker: 08:50 

So I'm going to give some inside baseball here and tell people a little bit about how we handle emergencies and problems, because people just don't know what we're doing behind the scenes. Somebody will send us a message in a number of manners saying, I can't print forms, I can't generate whatever it is. And what happens then is we try to validate it. But if it's especially if it's more than one customer, we go into slack. That's our communication tool and we have A911 channel.

My gosh, that thing lights up when there's a problem because it's not just one person. Everybody stops what they're doing, dives into 911 and they figure out, what can I do to help, right? You've seen it.

Shawn Bowen: 09:29 

Well, there's that first, there's that first moment of, oh man, A911 just popped up. And it's company wide. Everybody's feeling it. If you're in a meeting and we see a 911 pop up, everybody's aware of it. Yeah.

But then that activates, right? And it's really, I don't know if exciting is the right word or gratifying. I'm not sure what the right word to assign to this is, but there's a good feeling when you see everybody jump onto it in a way that is team oriented. Everybody knows their roles. The by and large, with.

It's A911. We're stopping what we're doing so we can address it, especially if we're immediately involved in it. And it's really neat to see that type of teamwork and camaraderie that comes out of those processes. Again, it's 911. So that can't be classified as something that we're excited to have happen.

But the way in which we approach and handle our nine one ones, I think, speaks well to the team and the culture that we've developed here at Quik!.

Richard Walker: 10:32 

Yeah, and we have evolved a lot in this process over the past five years because we're paying attention to what customers want so they can subscribe to status updates and get messages. And then when there's A911 and we validate it as a real incident, they're getting messaging. We're trying to be more proactive with them. You have implemented something in our company I didn't even know was possible. And I'm going to ask you to describe it because I'm not going to get the wording right.

But we have now gone to customers and said, tell us about your what do you call it, API contract. How are you. Yeah. Will we talk about that?

Shawn Bowen: 11:05 

Sure. So I just came on board with Quik! about a little over a year ago now. And one of the process improvements that was very near and dear to my heart was migrating to a true CI CD environment. The bottom line here is that CI CD just means that when code is committed after a software developer finishes their work and it's been peer reviewed, it gets deployed all the way out into production. Now, that probably sounds pretty scary, and it would be if we didn't have guardrails around that.

And so one guardrail that a lot of people may have heard in the technology space would be unit testing, for example. But that really just scratches the surface. And it only does one type of automated testing. The type of automated testing rich that you're referring to is called contract testing. And as an API company, our product is the API surface.

We call that the contract. It's the contract we make as a server to the client, which is our, our partners and our clients. That's what they're actually interfacing with. Now, as you might imagine, since these are connected software systems, the contract that we make is binding and that's why we use that term. If we were to change something about our API that would materially affect all of the customers that are using our software.

It's the quickest way to create a system wide outage, for sure. And so one of the things that you want to do in a CIC environment is, one, make sure that your contract is not changing. And we call that immutability. That's pretty easy to do because we can just check that all of the endpoints have the same name, the same URL. They all accept the same parameters and they all return the same data model when it's called.

So that part's relatively easy. But again, just like unit testing, it only tells you part of the story. That tells us that yes, the contract has remained the same and that when somebody calls into our API, the calls will be accepted. What it doesn't tell you, though, is what will actually happen when a real life client calls that API. And so what I think you're alluding to, Rich, is that the next step above contract testing is actual playbook testing.

And that is where we go out to our clients and we ask them specifically, how do you use our API? In what sequence are you calling these endpoints, and what are some typical payloads that you're sending to those endpoints? And what is your expected model that you're expecting to get back when you make those calls? So that allows us to not only do the unit testing we talked about earlier, but it allows us to make sure our contract is never changing. And the gold standard is then we're actually going to be calling into our production like environments with those very similar calls that our clients would be calling and getting the same results or getting expected results.

Now, if any of these steps fail, then that deployment I talked about from the developer's perspective fails and the pipeline stops and we don't have any deployment to production, so we don't have any failures. But then we need to remediate. We need to figure out why there is an issue, and then we need to send a patch up the pipeline and return to the CI CD environment. So long-winded way of saying all of these tools and techniques allow us to remain nimble and agile by having lots of deployments on a rapid basis, but it also gives us that safety and security to ensure that we're not breaking things inadvertently.

Richard Walker: 14:57 

Okay. I'm going to use a really bad analogy because I'm not even a coffee drinker, but this is kind of like saying if the barista makes a coffee your way and we change the machine in the background, she should still be able to make the coffee in the exact same way. Sure.

Shawn Bowen: 15:10 

Yeah. You as the customer shouldn't care, right? You should not care. Or. Or even the coffee beans.

As long as the coffee beans taste the same where they were sourced from. Now, I know I'm offending a lot of people because they're going to say there's no such thing as two different coffee beans that taste the same. But you're right, our customers shouldn't have to be concerned with how we're doing our implementation. They should care about the surface, and that's what they do care about. They care about the API surface, how they're calling in, and the expected results from that.

Everything that happens behind the scenes from there is a black box, but it's on us to make sure that black box is reliable and always available.

Richard Walker: 15:53 

Like being an API first company, I want to kind of spell this out for people because they may not really understand this. There's lots of APIs, their interfaces for how software communicates. But when we're talking about Quik!, we're talking about the ability to generate forms predominantly. Most people don't realize there's some 250 API parameters to generate a single form. So the combinations are infinite for what customers can do.

And it stymies us sometimes, right? We roll out something. I'll give you a case. We rolled out something where we fixed a bug, which just changed how the parameter was working, not not the value that went into it or not, but just how it worked. Because we fixed a bug.

What we didn't know is that a customer expected the bug.

Shawn Bowen: 16:38 

I know the exact. I know the exact scenario you're talking about. And that's absolutely part and parcel of being an API company. You just don't know how people are going to use your APIs. It's often going to be in novel and unique ways that we didn't anticipate.

That's why it's absolutely critical for us to not just monitor what we're doing, but monitor and understand what our clients are doing and what they're relying on. Because at that time, we didn't have a deep inspection of the customer's playbook. So we were unaware when we rolled this out, that this side effect that that customer was relying upon was actually going to be a big problem. We learned later, but that's exactly the way we don't want to learn is by the client giving us a call. You know, we want to be proactive about that.

And so yeah, that was a perfect example of one of those scenarios where if you're not actually checking what your clients are really doing and how they're really using your system, then you're only operating with part of the picture. And it's not a very clear picture at all in my mind.

Richard Walker: 17:47 

Yeah. Look, I hope my audience isn't bored with tech talk, but really elevates this. This is about understanding customer expectations and meeting those expectations consistently as your business evolves and grows and scales and changes. So let's turn to a more exciting topic. Shawn, you sound great.

We never have enough time to talk. You're in California. I'm in Texas. We just never have enough time to talk. And one of our favorite topics is AI.

Shawn Bowen: 18:15 

Oh for sure.

Richard Walker: 18:16 

All right. What is your view of how AI is transforming Quik!?

Shawn Bowen: 18:22 

Wow. In so many ways. And sometimes I'm just going to take a step back and say that I've been in professional software development, like Rich said, for actually 30 years. And I have seen throughout the course of my life three major changes that have impacted society and have been brought about because of the advent of technology. The first was the personal computing revolution, right in the early 80s.

I remember getting our first 286. It was a profound moment for me, seeing the possibilities that and the empowerment really that that this computing created. And then I got to see after college, the internet explosion. And we went from the idea that a person as an individual could do computing to we as a society are working together to do computing. And that changed everything, right?

Yeah. All the way from the rideshare and the gig economy to Amazon, and the downward pressure that places on brick and mortar institutions. I can't in my mind, I just think it was such a profound moment where our society changed dramatically, mostly for the better. I'm always an optimist, but it would be naive to say that there weren't downsides to that adjustment as well. Yeah, and that brings me to AI, because I believe wholeheartedly that AI is going to be as impactful, if not more impactful than the two revolutions that came previously.

And it's wild to think that this has all happened in my lifetime. So the good boy, it's a huge force multiplier and in very real ways, it's also an egalitarianism of software development. I'm going to give you just a brief anecdote that has nothing to do with Quik! and everything to do with my home life. My son is getting ready to leave for college. But in the meantime, he has developed this company called Noahtech Shameless Plug, where he does refurbishment of computer computers and computer-related hardware.

So for example, he'll buy lots from Newegg of, you know, 150 motherboards, and he'll go and he'll repair the ones that are relatively easy to repair, and then he'll resell those on the market and those that he can't repair. He'll just bundle those back up and sell those as a for parts or for scrap type of thing. So great. But where does AI come in? Well, he started looking at a lot of the software packages like QuickBooks and CRM software and all the other types of things that we typically cobble together to create a functioning enterprise solution.

And he had a really interesting Feedback to me, he said. You know, dad, I'm not really sure why I would use any of this when I can literally just create the software myself. And at first I was taken aback by that. I was like, you may probably go down this road and you're going to see very soon that no, these enterprise software packages, they're necessary. Well, and I'm not exaggerating over this last weekend, he put together an entire application.

It's a desktop application using react. And I won't get into the details, but it does everything. It does inventory management for him. It does his tax planning. It does it.

There's this website that he buys his lots off. It scrapes the information from those websites. It does the analysis based off of the SKUs and the pricing that he has, and his integration with eBay to determine whether or not these would be good deals. And then it ranks them accordingly. And then as he goes through and purchases them, it will actually keep track of his inventory, his profit and loss.

It's bonkers. And he did this without any programming of his own. Over the course of a weekend where he had, you know, probably 15 to 16 hours worth of free time. And so that's where I see the real power and the excitement of AI is it's empowering and enabling us to do things at an individual level that we weren't able to do before. Now, from a Quik! perspective, boy, we have just embraced AI in a way that is super exciting and also at times somewhat scary because when you're on the bleeding edge, it's hard to know whether or not your investment is going to pay off.

So Rich, as you know, we've spent the last year working on developing AI solutions, processes, and methodologies within our organization. And at this point, I'm starting to see real material gains in both our throughput and productivity, but also in the quality and consistency of the code that we develop. One of the things that I'm really excited about is the way in which AI can assist our programmers in actually creating best of class solutions. Some of our programmers are more junior than others, and for those that are less experienced, AI creates a really good tutorial for them for how to architect a system, a really good tight feedback loop. So they're not always requiring relying upon our seniors to come in and mentor.

It also gives them the ability to basically do some self mentoring as well through AI. We're also using AI in a very interesting way now with our pull requests. So for every yeah, take a step back. Every time a software developer submits source code, another of their peers is responsible for reviewing that. So we always have two eyes on any piece of software that's produced.

Well, we've now implemented AI based pull request reviews as well. So in addition to the humans that will always be involved in that process, we now also have an AI agent that is also reviewing it. And we've had some really great catches from the AI that our own peer peer review people didn't catch. So overall, I'm seeing it as a force multiplier. It's allowing our engineers to work faster and spend less time on the scaffolding and more time on the actual problem statement.

And it's also allowing us to ship code more safely by making sure that we always have eyes and ears on the pull request to ensure that when we actually do promote this into a deployed code base, that we're not falling into any sort of unobserved issues. So yeah, I've been really excited about it. There are some downsides to it as well. It's pretty easy, I think, for people to become overreliant on it and not apply their own critical thinking. So that's why I always caution the engineers to think of AI as a tool and assistant.

Maybe even an employee. But as an employee, you're the manager and you're responsible for their work. So you need to review it and you need to make sure that it aligns with all of our best practices.

Richard Walker: 25:46 

Yeah. You know, I didn't ask you the question directly, but you're, you're actually answering it, which is how your role in technology is improving customer experience. And one of the key things that you're doing is mitigating bad experiences. You're reducing the number of bad experiences.

Shawn Bowen: 26:02 

Look, we have a mature product. There are new products that we are producing, but Quik! is a 20 year old company. We have a mature product. It has market fit. It's a great solution Really from my perspective with with regards to that particular product, our core forms business, most of what we need to do is make sure that we're just providing the service that our customers deserve, which is uninterruptable, low latency, all of those type of features that you expect from an API.

In other words, you just forget about it. It just works. And so AI allows us to also do one of the things that we're working on right now actually is creating a solution where AI knows our source code base, not from a myopic perspective or just a file by file perspective, but more as an architectural perspective. So it can drill down through all the different layers of our service and understand where failure points might occur. Report on areas that we could improve upon.

 And in certain cases, we're even experimenting with small tasks, AI just completing the task for us. So it's an exciting time to be alive for sure.

Richard Walker: 27:18 

Yeah. Yeah. Look, I wanted to talk about this project actually, of creating documentation, AI based documentation for the purpose of AI against the code base, because anybody out there who's listening who has a five, ten, 20 year old company like we do, we're in 24 years now and you have legacy projects and you have supporting architecture that's old and you want to modernize it. The question becomes, how do you reinvent it? How do you modernize it?

And one of the things I want to point out to people is that AI is not just a bandage to be a force multiplier, you have to give it the right context. And sometimes that means doing projects and work to be able to lift that into context. So going through all our code bases and coming up with a model where AI can understand it and see it holistically is a foundational piece to let us go to the next level of modernization or speeding up tasks or training, training new employees, right? Yeah.

Shawn Bowen: 28:12 

Yeah. For sure. And I like the way that you describe it as, you know, when we're managing a person, they need to have context, right? And that's why we have started talking about AI as an employee or as a group of employees, because at the end of the day, if I were to hire somebody and just give them a task without any context, the likelihood that task would be completed successfully is just about zero. What I need to do is to make sure that I'm handling that same proper HR procedure of onboarding and getting them ready and prepared to actually do the work.

And so when we talk about AI, it's great to just go to ChatGPT or Cloud and type in a question and get an answer. But those answers are usually not necessarily tailored to you or your particular circumstances. They're usually generalized. Yeah. And that's great.

There's value in that, but there's more value in having something that is specific to your use case, specific to your situation. For that, it needs context. And so one of the projects that we are just wrapping up right now is a project where we're allowing AI to scour all of our repos and pull out information about that in. You might think of it as like an executive report about each one of those repos and all of the code that's contained inside of it. We made the decision not to allow, though, that information to actually contain the actual source code, and we did that in a specific reason, because what the end result of this particular project is, is a chat bot, basically within our public facing documentation that our customers can go and they can get answers from in a way that is more sophisticated than just this is the API signature, these are the parameters we're expecting, and this is the result you should expect.

This tool allows people to actually. Yeah.

Richard Walker: 30:10 

When I think about API documentation, it is boring. Nobody wants to read it. In fact, I've always said software has to be so easy, so easy to use. You don't need a user guide. And what I think about the future of AI, I think of dynamic websites that are chat based or interactive.

So instead of going to some guide and drilling down and looking for methods and then finding the right method and then looking at the proper, just ask it the question you have, how do I connect Quik! to DocuSign? Boom, here's how to do it. Here's all the data. Here's all the parameters, right?

Shawn Bowen: 30:40 

So we're just finishing up with this product. It's not quite live yet, but oh my gosh, I was tinkering around with it. And it's so much more even than what you just described. I can, for example, in the chat window say, hey, I want to call Quik! execute PDF method from let's say node. It will actually write a JavaScript implementation for you of exactly what you want.

And if you then say, oh, you know what? Actually, I really wanted to do it this way. It will then adjust that sample code, which is not really sample code anymore. This is actually just copy and paste it right into your application and it'll work. And so I find this to be. I'm hoping that this is well adopted.

This was a sort of an idea that I had late last year about ways in which we could improve our customer onboarding and maintenance. But also it helps with onboarding our own developers as well, because they can have deep conversations with the AI about the work they're doing. And it has a knowledge of our code base. So while that chat application was the first public facing exposure to our knowledge base that we're building inside this AI, it's certainly not the only application of it. of it.

In fact, I would say there's quite a few more impactful things that we're going to be doing with that knowledge base in the future. But building the knowledge base to begin with was, in my mind, the first step. And now that we have it, oh man, there's just so many opportunities for us to utilize that and make everybody just more knowledgeable about the domain that they're working in.

Richard Walker: 32:28 

Yeah. There's so many ways we could talk about this, but I want to point out something that I hope everybody caught. If you're going to spend three months, six months training an employee, why wouldn't you invest 3 to 6 months building out a context-rich platform to then train AI to give AI the tool sets the context and people as well. And then it expands beyond that to documentation and support and everything else. So I really don't think a lot of people put enough effort into the planning of how to use AI.

They're like, oh, it's fast. I can just use it.

Shawn Bowen: 33:03 

Right?

Richard Walker: 33:04 

But it's so much more powerful when you give it the fuel.

Shawn Bowen: 33:07 

We were just having this conversation earlier today too, about like, what the safe harbors are that we can give to AI and allow them to understand about our company? Because in my mind, the more AI knows about our company, the more valuable that AI becomes. But there's a fine line because there is certain information we don't want to be in the AI model, and we need to guard against that. So one of the things, for example, for a public-facing chatbot, we didn't want the actual source code to ever be leaked. Right?

That's intellectual property. But it extends beyond that from an internal perspective. We were talking about email, and I think we both agree that email is off limits. That's that private information shouldn't be part of the AI model. But then there's more question mark ones.

Should all our JIRA tickets be in there? Should all confluence be in there? Should all of our Dropbox documents be there? And it really becomes an interest. Yeah, no, it really becomes an interesting question about what is going to provide contextual value to AI and what do we need to safeguard against?

Because again, AI is a tool and we can't abdicate our responsibility to AI. We need to do so in a responsible way that protects not only our organization, but the individuals working for our organization and any of our partners or clients that work with us as well.

Richard Walker: 34:32 

Yeah, yeah, man, there's so many more directions we could go, but we're out of time. Shawn. So before I get to my very last question, what is the best way for people to find and connect with you and us?

Shawn Bowen: 34:45 

Yeah, for sure. So I'm on LinkedIn. You can find me under Shawn Bowen, Quik! CTO. And of course,