AWS for Industries

Convenient Conversations: How conversational AI elevates the traveler experience

This is a point of view from Andrei Papancea, Cofounder and CEO of NLX – an AWS technology partner specialized in building the next generation of conversational applications.

As consumer demands are changing, brands are constantly looking for new ways to interact with customers. Digital conversations are on the rise. According to’s State of Messaging Report, 83% of consumers contact a business over messaging to learn about a product or service, 76% to get support and 75% to make a purchase.

Multichannel digital servicing is no longer optional but expected. Surveys show that close to 60% of consumers don’t mind which channel they use to communicate with their bank, insurance company, or when they are seeking investment advice. Why would travel be any different?

As a frequent traveler, I often find myself spending too much time on routine tasks such as booking the right flight, hotel, or rental car, but it doesn’t happen for the reasons that you might expect. Yes, budget constraints are there, but they are not significant enough to be the issue. For instance, my company’s policy is to fly economy class for domestic flights. It does not state the maximum amount that an employee can spend on an economy-class flight, nor does it restrict the airline carrier.

At the same time, the bulk of my time is spent on filling and refilling the same forms over and over again, then slicing and dicing through the results in a quest for the same segment almost every time: an early morning economy class flight leaving LaGuardia or JFK, via one of my two to three preferred airlines, a window seat, a four-star (or up) hotel that is close to downtown, and an economy rental car that is picked up and dropped off at the airport.

Around travel time, I always look for a last-minute seat upgrade, a rental car upgrade, and an upgrade to a room with a view. It would be so much easier if I could just say where I am flying to, on which dates, and then receive a travel itinerary that takes into account all my travel preferences and simply allows me to book the end-to-end trip.

Being the CEO of a startup company, the purpose of most of my trips is to meet and listen to customers – many of them from the travel and hospitality industry. The common thread across all customer anecdotes I hear is to not boil the ocean and instead try and solve the last inch of friction during the trip lifecycle. Technology already provides an extremely efficient way that allows travelers to move around the world in a seamless fashion.

Customers are looking for a common layer that works when other things don’t – a layer that provides a connected experience across the reality of a fragmented travel landscape. Having worked for a large travel management company, I’ve seen how even slow seasons can hit call centers with unexpected difficulties. Take severe weather, for example, and they come to a standstill. A challenge in scaling customer service that does not exist for an intelligent conversational platform. On the other end of the spectrum is the connected experience for “markets-of-one” or hyper personalization. As a traveler, I want to be able to receive or request a relevant recommendation at the right moment through the right channel. I believe that this convenient future is almost here and that it will be brought to life by AI-powered voice applications.

Possibilities of conversational AI for travel and hospitality

Is the “AI-powered-smart-travel assistant” making a comeback?

“Travel is an industry that’s taking a leading role in the adoption of modern messaging to elevate experiences throughout a customer’s journey.” says Warren Levitan, VP Conversational Business, Zendesk.

Conversational commerce has hit the mainstream media in travel and hospitality, but my perspective is it has not yet found the adoption it could have. The real evolution has been modern cloud technology, helping to advance the access to and affordable experimentation with AI/ML services to drive innovation fast and deploy rapid improvements in production.

Imagine a voice-enabled platform that interacts, listens, and learns at every stage of the travel process, from travel intent to travel expense, re-booking or upselling, destination content or duty of care. This is how I envision we bring ultimate convenience into the ecosystem.

This convenience is to be powered by conversational AI, which as a generic channel of distribution, entails any text- or voice-based application that offers value through human-like conversation. Conversational AI is a great enabler for both multichannel digital servicing and personalization, for much lower costs and faster time to market than traditional channels. You no longer need your own website or mobile app to reach your customers. All you have to do, instead, is to distribute your applications and services over well-established channels such as phone calls, mobile messaging apps, or smart home devices such as Amazon Alexa. Furthermore, companies can deliver state-of-the-art personalization in a matter of minutes using services like Amazon Personalize and Amazon SageMaker, with limited to no knowledge of machine learning required – accelerating time to market even further.

Amazon Alexa Skill for airlines

In practice, the best candidates for voice applications are use cases that can be simplified by reducing the number of steps required by their current implementation. A good example is flight check-in.

Most airline check-in flows today require customers to:

1.     Open their phone or laptop

2.     Open up their email client or the airline’s mobile app

3.     Click through multiple screens to actually check in

4.      Add or text the boarding pass to your phone for use at the airport

A much simpler and natural flow is, “Alexa, check me in for my flight tomorrow,” to which Alexa asks a security question – plus possibly prompts a last-minute upgrade offer – and then texts you the mobile boarding pass. All of this is done hands-free while in your office, car, or the comfort of your home.

The company I cofounded, NLX, is partnering with AWS and our customers to bring such experiences to travelers everywhere. One of our recent developments is a managed Amazon Alexa Skill that airlines can turn-key to market and distribute to their customers in minutes. You can find the reference architecture on the AWS Architecture Center.

The Alexa Skill enables travelers to check in, book seat upgrades, check their flight status, and more, all done seamlessly via voice. What makes this solution special is how easy it is for airlines to take an integrated experience to market, without the need for development or technical expertise – an industry first. Our solution relies heavily on serverless and managed services from AWS such as Amazon API Gateway, AWS Lambda, and Amazon Elastic Container Service (Amazon ECS), which ensures that our systems scale automatically with our customers’ needs.


The fast pace and quality of our innovation is fueled continuously by our partnership with AWS. From cutting-edge infrastructure and managed services, all the way to the unmatched support, AWS enables NLX to build and deliver our market-defining technology at a fraction of the cost and speed of traditional systems. The end result is simple yet powerful: our customers do not have to think about the underlying technology, but instead, they can focus on what matters the most – their customers’ experience.

We partner with AWS because they share our deep passion for innovation on behalf of our customers and allow us to democratize conversational AI technology by making it more accessible for developers to use without programming.

I anticipate significant innovation driven by conversational commerce between suppliers and travelers, which allows us to reach a point of true convenience in the travel ecosystem. On that note, my prediction is that by 2022, more than 50% of travel will be booked and managed at least in part via voice, bringing more convenience and efficiency to travelers all over, which in turn will lead to higher levels of customer loyalty and satisfaction.

Interview: CEO in the spotlight

Florian Tinnus, Head of Global Partners for AWS Travel and Hospitality, talks with Andrei Papancea about his experience as a CEO of NLX.

Florian Tinnus: NLX’s tagline is “let your data do the talking” – can you tell me more about this?

Andrei Papancea: From the very beginning, NLX’s purpose has been to enable customers to leverage conversational AI technology without having to program. In other words, NLX’s goal is to empower customers to deliver cutting-edge, AI-powered conversational experiences, without having to worry about their underlying implementation – and what better way to make an experience stand out, if not by personalizing it with data? From ideation and experience design, to data integration and deployment, NLX helps companies deliver data-rich customer experiences in a matter of minutes. So, in a nutshell, don’t worry about the technology, just “let your data do the talking.”

FT: What keeps you up at night?

AP: A thousand different things, but most importantly: speed and agility, as we innovate with and on behalf of our customers. In a crowded space, NLX differentiates itself through the ability to deliver complex and robust conversational applications at a rapid pace, which could not be made possible without AWS’s advancements in cloud technology and infrastructure.

FT: What are you most excited about in 2020?

AP: On the NLX front, I’m very excited about a collection of new features coming out in Q1 that are meant to push the boundaries of what is possible with voice AI today. On a separate note, I’m excited to teach a new course at Columbia University, titled “AI Engineering.” The course focuses on taking AI model development from your laptop to operationalizing it in the cloud, with the end goal of building large-scale, AI-powered systems that can be used by millions.

Learn more about how AWS is enabling the most innovative Travel and Hospitality brands. 

Andrei Papancea

Andrei Papancea

Andrei is the CEO, chief product officer, and cofounder of NLX, a leading customer self-service automation solution. Its conversational AI software-as-a-service (SaaS) products help brands transform their customer interactions into automated, personalized self-service experiences. NLX is an AWS Travel and Hospitality Independent Software Vendor (ISV) Competency Partner, one of eight AWS Conversational AI Partners, and an Advanced Technology Select Services Partner. Prior to cofounding NLX, Andrei built the Natural Language Understanding platform for American Express (AmEx), processing millions of conversations across AmEx’s main servicing channels.

Florian Tinnus

Florian Tinnus

Florian Tinnus brings 20+ years of experience in the travel and hospitality industry with leading commercial roles for IT solutions and strategic partner management at Fortune 500 companies and technology players. Based in Boston, MA, he is responsible to lead the AWS Partner Ecosystem for Travel & Hospitality globally. Florian holds a degree in business economics and is a certified airline commercial manager. He lived and worked in Germany, France and the United States with a passion for all things travel and technology.