Andrei Papancea, Co-founder & CEO at NLX.ai talks about AI chatbots, smart assistants, smart IVR applications, and multi-modal experiences shaping the tech experiences.
1. Tell us about your role at NLX. How much of your typical day is involved in innovating AI tech for your customers?
I’m the CEO and Co-Founder of NLX, which I started back in 2017 with my brother and a middle-school friend. We’re all passionate builders and in NLX we found a great opportunity to satisfy our interests, while pushing the boundaries of the conversational AI space. While I spend a lot less time touching code nowadays, I find myself brainstorming new ways of delivering our products and services quite frequently. Our approach is simple: always walk backwards from the customer problem and assume anything is possible. Following this approach, innovation is always part of the solution.
2. What are the applications or rather opportunities you seek to have with your product?
Our product offering is fairly comprehensive for use cases across the board: from AI chatbots to smart home assistants, and from smart IVR applications to multi-modal experiences, our products can deliver almost any conversational experience. Our constant pursuit is to come up with new ways to simplify the building of these applications and power conversational experiences that people want to use.
3. How did you define the vision of NLX? How did you approach your first 100 days at NLX?
Our initial vision was to build a platform that allows companies to build and manage conversational AI applications across all channels, at enterprise scale. This vision emerged from my experience building such applications for companies such as American Express and other Fortune 500 companies. Given that I’ve done this before, the team and I jumped in head-first and built the initial version of our NLX Studio platform.
4. What are some of the unique lessons you have learnt from analysing your customer behaviour?
First and foremost, one of the main lessons is that you cannot predict human behavior when you expose a conversational application to end-users.Offering these AI applications to your customers is almost like putting up your own corporate “Google” — where people will feel free to ask all kinds of things in many different ways. The key to this problem is a great learning in itself, as the solution is to not necessarily try to predict what the user will say, but instead to be prepared to analyze and extract insights from what is said. For instance, identifying trending topics in customer conversations is an extremely useful way of deflecting traffic from contact centers. With the right tooling, you can pick up on trends after the first handful of people request assistance for such new topics, which will then allow you to train your AI models to detect the new topic and deflect the subsequent traffic.
5. NLX recently achieved AWS Travel and Hospitality competency status. Can you tell us more about the same? How it will help the customers?
AWS launched the AWS Travel and Hospitality Competency to help customers find highly specialized AWS Partners to streamline cloud computing adoption within the Travel & Hospitality industry. NLX is proud to be one of the 27 launch partners of the new competency. This designation recognizes that NLX has demonstrated deep experience helping customers transform their business from behind-the-scenes operational efficiencies to guest-facing customer experiences. The launch of the AWS Travel and Hospitality Competency comes at a crucial time for the industry as it looks to innovate to address changing demands in the face of COVID-19.
6. What are some of the common pain points that your customers commonly approach you with?
Over the past year, the most common customer pain point has by far been trying to meet end-user demand in contact centers. The pandemic has exposed a lot of operational issues for companies, the majority of which have to do with the challenge of scaling customer service on demand. In line with market trends, our customers are also increasingly experiencing younger users who do not want to talk to a person to get something done. They expect their customer services to be self-help, fast, personalized and accurate.
7. What advice would you like to give to the upcoming Conversational AI-based tech start-ups?
Iterate in small increments and collect lots of customer feedback along the way. This will allow you to stay nimble and pivot based on the customer demand. Also, make sure to think through your system architecture early on. Building AI systems is 20% model building and 80% systems engineering. You don’t want to be stuck with a nifty AI model on your laptop, so you need to make sure you plan for scale early on. Lastly, don’t underestimate the superpower that cloud services can provide for your company. At NLX we benefit from unmatched development agility when you pair our industry expertise with the power and breadth of AWS.
8. Can you give us a sneak peek into some of the upcoming product upgrades that your customers can look forward to?
Let’s just say that our Voice Compass technology can operate through any bi-directional audio channel. We started off with a call center implementation of Voice Compass because it’s the most commonly used method of servicing customers today. At the same time, we are seeing increased demand for support in non-traditional channels (related to training and DIY instruction) that we see as a strategic way to capture more market share in the future.
9. Which is the one AI breakthrough you will be on the lookout for in the upcoming year?
I’m on a constant quest for breakthroughs in the smart home space, specifically innovative new ways to make smart home AI assistants more useful in our daily lives. Specifically, going beyond switching lights on and off, playing music, and getting weather information. With the ever-growing presence of these devices in households across the world, any breakthroughs in how and what we interact with these AI assistants about, have the power to forever shape the way we interact with computers moving forward.
10. Please share a recent piece of content (can be video, podcast, blog, movie, webinar) that resonated the most with you with respect to Conversational AI or your work in general?
I found this announcement from Facebook earlier this year to be a pretty awesome breakthrough. Their M2M-100 model allows translation between any pair of 100 languages without having to translate into English as an intermediary step, which used to be the norm. Best of all, it outperforms previous English-centric systems by 10%.
11. What is your go-to strategy that brings technology and the human side of marketing together?
The strategy that has worked for us really well in the past year is to have a well-crafted, user-centric demo that people can try out for free. This invites people to try out our technology product without any hard commitment, while at the same rewarding them for their time with a fun, tailored experience.
At the end of the day, what is consumed is an experience not a piece of technology.Whereas mobile applications had hardware, visuals and interaction, voice applications are not seen – they are heard and experienced. Allowing easy access to that experience is the key that brings technology and the human side of marketing together.
12. What is the one quote that has stayed with you throughout your professional life
“Believe you can and you’re halfway there”. This quote from Theodore Roosevelt fully embodies how I’ve approached every problem or challenge for as long as I can remember. It is also the guiding principle for everything we do at NLX.
Co-founder and CEO at NLX.ai
Having previously worked for American Express, where he built their conversational AI platform from scratch, which they continue to use to handle and automate their chat volume on both their website and mobile app, Andre co-founded NLX in 2018 to provide a comprehensive approach to building and powering conversational experiences.