Category: Podcast

Introduction to Guest

Jonathon Morgan is the founder and CEO of Yonder, an AI company that helps Fortune 500 communications teams identify and counteract online disinformation about issues that matter to their organization. Prior to Yonder, Jonathon developed AI to combat social media
radicalization with DARPA and served as an advisor to the US State Department. Jonathon has published research about radicalization and online disinformation with the Brookings Institution, The Atlantic, and the Washington Post, and has presented findings to NATO’s Center of Excellence for Defense Against Terrorism, the United States Institute for Peace, and the African Union. His op-eds about how social media platforms and policymakers should address the threat of disinformation have appeared in CNN, The Guardian, and Slate.

In Jonathon’s early career, he spent a lot of time thinking about and developing communities on the early social internet. In fact, Jonathan was one of the first professional bloggers, which was a unique profession at that time. Through his blogging, he created a small, but passionate, audience, all deeply engaged in his work. In the early days of social media, even before Twitter existed, some of the larger social media platforms started to take off, and there was this newfound wisdom. As Jonathon developed professionally which included really understanding the dynamics of the internet, he embarked on a new path that led him deep into the technology world.

Jonathan was a software engineer for a while, working as a data scientist. Because he had diverse backgrounds both in social media and software engineering, he ended up in this interesting spot where he understood how these social dynamics were changing, especially as more people connected on social media over shared passions and interests. Jonathan also understood the dynamics of social communities in a positive way, and how these communities came together.

Jonathan was able to understand larger dynamics at scale by thinking about tech approaches. For example, back in 2013 when extremists’ groups like ISIS used social media to hijack the public’s attention and spread propaganda. He realized the way society communicates was greatly changing – and there were many consequences involved with these massive changes as well.

Coming Up with The Idea

As mentioned previously, the idea behind Yonder started with the rise of ISIS. There was speculation of ISIS having millions of followers around the world, and Jonathon and his friend (a counterterrorism researcher) decided to try and figure out how big these follower groups really were. What they discovered was there were maybe 30k – 40k accounts that were supporting the ISIS agenda. ISIS used a small number of accounts and automation coordination to create the impression that there must be millions and millions of followers. The idea that you could take a small group and coordinate energy, attention, focus and effort to create the impression of a larger group on the internet was a massive lightbulb moment for Jonathon and the team.

Flashforward to 2015 and 2016. Jonathan and the team were witnessing similar dynamics associated with domestic radicalization in the US, and through large, seemingly coordinated information campaigns. It was clear that they were being orchestrated by a single source or coordinated in some unique way. It was also clear that anybody who was pursuing an agenda on the internet could use these dynamics to accomplish a goal. These groups were proving to have as much power and influence in a conversation as a global corporation with a limitless marketing budget. Jonathan knew this concept would make an excellent technology company, especially when addressing larger problems in our information ecosystem – such as misinformation, conspiracy theories, etc. Soon after Yonder was born.

Testing The Product & Validating the Market

Jonathon felt like he had created novel technology that was applicable to a very large problem – stopping the spread of misinformation and online bullying/harassment. Jonathan and his team thought – who would use this product? What is their workflow like? What problems are they solving? What is their job? They came to the conclusion that this problem was very important, and Yonder had the expertise and technology to make an impact. Yonder was soon able to find some early customers who shared the same concern about this online problem, which led to investors interested in the company. Yonder realized that you must develop something that people will use to make their jobs better. This was a big transition for Yonder. Jonathan and his team knew they wanted to pursue this online problem of subgroups and misinformation, and doing so meant partnering with customers to help their companies.

Capturing The Misinformation

Yonder is focused on specific social media sites: mainstream platforms and Twitter. What Yonder is really looking for is groups of accounts behaving the same way, almost working together, sharing the same agenda. This is the kind of technology underpinning that Yonder collects. Yonder then performs a network analysis, using natural language processing to bundle emerging narratives. Yonder identifies semantically similar content within a given window of time. This is what allows Yonder to capture this concept of a narrative on social sites.

Using The Platform

The Yonder platform is designed to identify coordinated groups. The platform does not have an opinion on whether the groups are good or bad, it just knows that there are networks of accounts that work together, that kind of move like schools of fish on the internet, working hand in hand to pursue an agenda. The Yonder platform helps identify these groups. What is valuable for Yonder’s customers is there’s always topics of conversation that are relevant to their brand. Maybe speaking about corporate values. Or if they have a product, the company needs to understand who is passionate about it, or who is opposed to their product, etc. The Yonder product can identify emerging narratives that are relevant to their communication objectives. Yonder is working with public/government affairs, corporate communications, PR, and other strategic communication functions that are increasingly taking on a ton of responsibility inside of an enterprise to protect brand reputation while staying true to brands’ values.

The groups mentioned need to be close to emerging narratives that matter to specific issues. Because you must align everyone internally, you must make sure that all the people who are spokespeople or representatives for the brand use messaging discipline. All these decisions can be hard without data. Yonder also discovered that by focusing on the agenda driven internet, Jonathon and his colleagues were able to identify narratives well before they go mainstream and/or covered by journalists. Yonder provides the additional time that allows these teams to be successful in an environment that otherwise is just based a lot on intuition. They are always trying to figure out what is real and what isn’t? Whose agenda is behind this? Is this going to blow up? Or is this just going to be a quiet story that goes away? Yonder helps companies evaluate all these decisions on a daily and weekly basis.

Challenges Encountered

One of the biggest challenges was zooming in on the technical issues early on. For example, finding the right way to engage with customers acting solely on intuition, and helping them enhance this intuition using the right (and correct) data. Making sure the data is valuable. Another challenge is how to predict 100% of the time when something is going to go viral on the internet. Yonder began looking into too many things at once, instead of just trying to figure out how to solve user and customer problems. That disconnect caused Yonder to stumble early on. However, once Yonder began to commit to strategic communications the company began to understand what their customer’s problems were, what they were trying to accomplish, and what really kept them up at night. Yonder also discovered that a simpler version of their technologies (a simpler product) would be more helpful. This has been an important learning curve for the company.

Advice For Brands Interested in Learning the Narrative

For brands that are learning the narrative of the internet, what Yonder has been seeing is that the role of strategic communication teams has become more prevalent and very important. The strategic communication teams are receiving a lot more responsibility and influence within their companies. Yonder believes this will be a trend for brands as they think about communicating their values and message, all while connecting with stakeholders.

A cool thing that strategic communicators do is find someone who has influence in a particular community, perhaps an industry thought leader or someone in the public eye, but who also has trust and respect within the community. They are aligned with the issue at hand. Thinking about coalition building online means really engaging with and thinking about these online factions. Over the next ten years, there will be huge changes in the way that brands think about communication and messaging strategies. And how they engage with the ups and downs of these highly motivated and focused internet subcultures.

Yonder is currently working with some cool, innovative brands that understand these dynamics and are eager to start making necessary changes. These brands are really looking at how to structure their organizations which include building teams with a keen idea of how online coalitions work.

Advice For Aspiring Entrepreneurs

As you think about raising VC, it really matters that you find a partner who’s going to be with you through the ups and the downs. There will be times where the business will really struggle. At that point, you will have to answer hard questions and make tough decisions, which is why you must have partners who trust you, believe in you, and will challenge you.

Taking on early founders will shape the trajectory of the company in many ways and will certainly shape your experience in the business. Make sure you evaluate the long-term impact of your company when looking for funding, for this partnership is like a marriage. Afterwards, think about how much you’re raising because every time you raise money, you raise it at a certain valuation. And venture capitalists like having an expectation that it will go up by a certain amount. So every time you raise money, the value of your business must be twice as large, or three or four times as large. Really understand and think about what you’re signing up for. Thinking that you just need as much cash as you can get your hands on might not be the best approach. Make sure you’re methodical about your decision and what is truly best for your business.

https://www.yonder-ai.com/

Introduction to Guest

Marc Stevens is the President & CEO of Zorroa, a Gradient Ventures-backed machine learning-startup that has recently launched Boon AI, a machine learning integration platform that is modernizing media management pipelines for the media & entertainment industry. He is a global technology executive with three decades of experience driving technology and business transformations across film, television, and gaming industries. Prior to Boon AI, Marc led the media & entertainment divisions at organizations including Autodesk, Avid, and Microsoft, where he drove M&A initiatives and incubated award-winning solutions like Maya, Flame, & Shotgun

Marc began working in the technology/AI space 20 years ago, later transitioning into consulting. Through his contacts, Marc was hired as a consultant for Zorroa, which was recently transformed into Boon AI. This began his relationship with the company. Zorroa was seed funded by gradient ventures. Marc met with the investors and discussed some ideas on how they could transform tech and business to become more custom development focused with a broader market appeal and scale. Marc’s vision was to get the technology out to more people while keeping the business scalable.

Coming Up with The Idea

Boon AI started out as a raw set of media technologists who came out of the film/visual effects space, with experience working on the largest, most complex, media projects such as: The Lord of the Rings franchise and Avengers movies. When the company was first founded, there was a lot of innovation in the AI machine learning and computer vision space – with many ideas on how to successfully leverage AI ML technology to transform digital media management. This is where the journey began. However, every customer engagement wound up being a custom development effort, which was not scalable in the long run. Marc and the team began to look at this from a different perspective. Was there a way they could start to put things in a form factor? Make the product more accessible? More affordable? This vision is what birthed Boon AI just a few short months ago.

Product Development Journey

Boon AI is the no code, machine learning integration platform for media driven companies. The product solves a couple different problems.

When Boon AI is creating machine learning products for clients, risk is reasonably high – for the results are not necessarily known from the start. Nor is it guaranteed that you will get the results that you want. There is data science behind creating a model that works, but then deploying that model and scaling it in production is another set of technologies. What Boon AI tries to do is package up some of the infrastructure parts, leveraging cloud technologies and off the shelf tools from Microsoft, Amazon, to Google. From there, tools are added to customize at a high level, with a nice visual interface that abstracts having to set up the whole infrastructure to even run one model on one image (which is time consuming within itself).

Within an hour, you are pointing to a bucket of images or videos, and depending on the program you’re trying to solve, you can go in and select the models you want to run. And there’s a nice visual interface to be able to understand and experiment with results. Boon AI tries to give people results that they can start to understand while experimenting with different options, and/or doing AB testing. Once a client knows what works for their use case, Boon AI can lock in that platform, built in a way that is scalable and able to handle large payloads.

A Common Use Case for Media Companies

Boon AI tends to be a bridge for companies to be able to get enriched dynamic data. Some companies like tagging information in an image, for example, so they know which images have a certain logo show up in them, or which images have a certain person they’re looking for. (For example: Robert Downey Jr.in a Lamborghini in NYC).

With Boon AI’s technology, everything will automatically pop up on the frame instantly, think usage rights management and content moderation. Another example is if a retail company wanted to see all images that look like a particular shirt. Boon AI’s product can find images like the shirt.

Customers & Target Market

Boon AI started branching out to companies that were connections in the media space, mostly production companies, (full length features, commercials, etc.). Boon AI is building catalogs for companies, so all their goods are online, making them searchable and accessible.

Boon AI is also working with sports teams. Many sports teams are looking to monetize their assets in different ways – like archived footage for example.

How the Software Works

Boon AI removes all the heavy lifting of setting up the infrastructure, allowing people to focus on the results to enhance their business. There is a subscription fee to join the platform.

Once you sign up for Boon AI, you create an account. Then Boon AI’s user interface walks you through a process of content you want to analyze. Depending upon what you’re interested in, you are presented with different choices of AI machine learning models that could be run for you. From there, you receive a nice visual display where you can start to search and filter from – do some AB testing against different models from different vendors. This whole process takes under an hour.

Challenges Encountered

One of the biggest challenges has been where is Boon AI’s identity as a company? Customers have pulled Boon AI in different directions. Is Boon AI a tool for developers that is in their toolbox, that they use to help get their job done? Or is the company going to provide an end user solution that does everything for PA? Will Boon AI fall somewhere in the middle? From a product market fit point of view, this has been one of the hardest challenges for Boon AI – finding their true identity. Another similar challenge is balancing being clear about what problems the company is solving versus how to solve them.

Advice for Aspiring Entrepreneurs

Focus is incredibly important. Have a vision of where you want to get to but have flexibility on how you will get there. Know that business fundamentals can change. Also, be prepared for customers not to be ready for your idea right away. It could be a fantastic idea but could take time for customers to see the company’s true potential.

Advice for Industry Leaders

See if you can create a team of people that might be able to give you more options, or who think about different ideas faster. 90% of AI/ML projects never make it into production. People have been bitten in the past, and worried about going back into the AI space. Think how your tools and technology will allow employees to spend time to add real value in your business.
Your employees should take work from your plate, while you provide tools and capabilities to assist them in being creative and successful. Focus on the creative pieces that work, the pieces that the machines are not going to be good at.

Another piece of advice, and something that Boon AI is focusing on, is to bring the fear of experimenting down. Know it’s not a big investment in risk but allows people to experiment and see where/how the products help their business.

https://www.boonai.io/

Introduction to Guest

Gideon Mendels is the CEO and co-founder of Comet, a leading provider of machine learning operations solutions that accelerate getting machine learning models into production. Before Comet Gideon founded GroupWize where they trained and deployed over 50 Natural Language Processing models on 15 different languages. His journey with NLP and Speech Recognition models began at Columbia University and Google where he worked on hate speech and deception detection.

Gideon started his career 15 years ago as a software engineer. He soon enrolled in grad school at Columbia and became more intrigued with machine learning, in particular speech recognition. Gideon built a system that was able to collect data, but it was tricky to identify which language it was in. This started his journey of learning natural language processing and language identification. Gideon stumbled upon a problem that was hard to solve with traditional software, and built his first model which was a language identifier

Gideon and his co-founder decided to start a fun side project building an app, and began by collecting data from WhatsApp, Viber, and GroupMe (for example). Through analyzing their findings, they were able to provide interesting metrics. Once released, the app blew up very quickly, hitting 80k users in the first week organically. Gideon knew he found his knack.

Gideon soon started a job at Google, working on hate speech detection on YouTube comments. He brought a lot of experience with document classification, which is essentially the technical term for hate speech detection and was able to improve the production of his team. This was a huge learning experience and time of growth for Gideon.

Coming Up with The Idea

Gideon and his team saw inefficiencies in how the operations around machine learning and data scientists were working in a very old-fashioned way. In these discoveries, they decided to interview over 150 data scientists, asking them about their workflow – How do they do things? How is it tracked? Etc. After this was done, Gideon created two similar products with slightly different variations. Unsure which product was best to go to market, Gideon decided to test both. He and his team bought two domains, built two landing pages, and then drove traffic to the landing pages. From the tracking, he realized ML was the better variation, which turned out to be what Comet is today. Basically, they tracked mock-ups of essentially a very, very early version of Comet.

Testing the Product & Validating the Market

Comet did plenty of market research, as mentioned above. What existed back then were companies and startups trying to solve all the problems in the machine learning workflow. Experimentation, management, model management, orchestration, deployment, monitoring, etc. Comet looked at these companies thinking the scope was way too large for one software company to build. So, Comet decided to do only one thing, and this one thing will be done better than anybody else. This was Comet’s approach.

Comet did talk to many potential clients through introductions, but the focus was a grassroots approach. After a week of launching the product, Comet received emails from data scientists and machine learning engineers working with big companies, loving what Comet had to offer. 90% of Comet’s enterprise customers started as community users.

Comet is a machine learning platform. It could easily be self-hosted by the customer or cloud base. Through working with data scientists and data science teams, the platform allows them to track, compare, explain, and optimize their experiments and models. The product solves problems around reproducibility and being able to help teams get to production faster with a successful model.

Comet now supports some of the biggest and best machine learning teams in the world within the healthcare, tech, media, and finance space. Gideon and his teamwork with Google, Ancestry, the National Institute of Health, to name a few, where they use Comet as their main development platform to build AI models.

One of the biggest challenges was finding the product market fit. While many other companies are doing similar things, Comet is really growing with the market, shifting, and changing by the day, so they are not replacing any legacy systems. From the product market fit perspective, Comet wants to be able to build a product that people are happy about and willing to pay money for. But one must think ahead, for the market is shifting all the time. And if you solve a pain point today, then that pain point is not going to exist in a year, because some of the other market dynamics are changing. This is super tricky. Discipline is key.

Future Endeavors

Comet thrives to be the de facto operating system for machine learning within enterprises while helping their customers bring business value with Artificial Intelligence. The company is very much focused on business-driven missionary teams rather than grad students and centers of excellence.

Another way to think about the future is when you walk in a room of software engineers, there’s usually a TV showing your build system and another TV showing your servers in production. Comet’s vision is in the next few years to have another TV showing their experiments and models. This will all be driven by content.

Advice for Aspiring Entrepreneurs

If you have a few different approaches and you do not know which model will work the best, see how you can try and model something quickly to see which product works the best. Building a successful company also comes down to the people. Comet hired people slowly, to really make sure the best suited candidates joined the team. Once you build that core team, you can start accelerating very fast. Do not compromise the quality of the people that you bring onto your team because who you hire will make all the difference.

https://www.comet.ml/site/

Introduction to Guest

Peter Ma is Co-Founder at MixPose, a live streaming yoga platform that uses AI pose detection in real time. He’s part of Intel Software Innovator, Arm Innovator and Nvidia AI innovator. Peter has been a software developer for more than 12 years, and has been involved in five different startups, not to mention winning more than 100 hackathons. Some of his accomplishments include speaking about mobile apps at TEDGlobal 2010, Winning the 2015 AT&T Developer Summit grand prize, and winning the grand challenger award of the United Nations Development Programme COVID-19 challenge (to name a few). Peter has previously founded Clean Water AI and Doctor Hazel, a skin cancer AI detection system. Peter graduated from the NJ Institute of Technology with a degree in Computer Science.

Ever since Peter was young and playing with computers, he has considered himself an engineer. He went on to receive a Computer Science and Electrical Engineering degree, began working at a good company, and soon found himself entering in, and winning hackathons. Eventually, Peter quit his full-time job, mainly because he started to make more money at the hackathons. Through participating in hackathons, Peter earned short term contracts with different companies building prototypes, while building his network and community of people. Because of these opportunities, Peter found himself surrounded by cutting edge technology. For example, Peter began working and learning about AI before it was widely talked about. Because of Peter’s natural curiosity and interest in technology, he saw each challenge and work opportunity as a chance to learn and grow.

Coming Up with The Idea

Peter and his friend began bopping around ideas with each other on an AI company they would like to start. With Artificial Intelligence able to detect body parts, they thought it would be a great idea to start MixPose – that can detect human poses in real time.

Peter then submitted MixPose to an Android challenge (hosted by Google), and his app was selected as one of the ten apps who won the worldwide competition. In fact, Peter and his team were the only team selected from the United States. After winning the competition, Peter and his team began getting to work, with the vision to make MixPose a highly successful and usable app. By January 2020 Peter and his team had put in a lot of effort into the company, however, were skeptical if people would teach or take classes online. However, after COVID hit, all Yoga classes were forced to be online only. This was the moment when Peter realized they had a product that could really work.

Developing the Product

MixPose is getting rolled out in three phases. Phase One is getting the product ready, and ready to produce the highest quality content. MixPose does not use Zoom, but instead a broadcasting software system along with music encoding. This way you are encoding movie ads in real time. Phase One also involves the teacher onboarding. Second Phase is the subscription. Third Phase is the independent classes so the teacher can teach and then get paid. MixPose generates income from subscriptions.

The AI comes into play if the teacher has a large crowd of about 100 students. The teacher can then use Artificial Intelligence to see if anyone in the class is doing the pose(s) incorrectly. That student will show up on the screen as high priority so the instructor can give the individual more attention. The instructor also has access to crowd analytics.

Right now, MixPose is testing the product out with different teachers, with about six to nine students showing up for each class, all of which have the option to be on or off camera. MixPose has a privacy option so you can join a private class so only the teacher can see you. You can also join friend’s groups. MixPose is one of the first apps to broadcast high quality video content as well.

A MixPose class versus a Zoom class is very different. Peter is teaching the MixPose instructors how to use game streaming software, so they are up to date with the technology.

Advice for Aspiring Entrepreneurs

Make sure to make a product demo and look at all start-ups from a business perspective. Go through the user-research, which Peter believes might be the most important. The goal is to really build your relationships over time, and to build a team you can trust. Trust people with money because they trust you with money. Really understand your teammates strengths and weaknesses so as you grow your company your team grows with you.

https://mixpose.com/

Introduction to Guest

John-Isaac Clark, known as “JC”, is the CEO of Arturo, an AI property analytics company that helps Insurance Carriers improve the accuracy and speed of decision making across claims, underwriting, pricing, and renewals. JC possesses more than 10 years in geographical and location-based analytics with a start-up background and experience. Formerly, JC was the head of Commercial Product at DigitalGlobe and considers himself a recovering software engineer.

JC’s journey started at a very young age as a self-taught programmer. Since JC was eight years old, he was always fascinated by computers and technology, especially since his father was in the industry, before there really was an industry. When JC was only 19, he wound up joining a startup in Chicago, leaving university behind to become an entrepreneur. The startup had some success, which led him to work in San Francisco and Austin for a few years. JC also worked through the .com era, which was an experience within itself. This time in JC’s life gave him a lot of exposure to being an entrepreneur and the startup culture in general. He started to ask questions like “Why are we building this?” and “What is the value we are creating with this?”. Through his experiences, JC found his spark as a Product Manager. Overtime, through one of JC’s last startups, T- Sciences , he worked closely with Google Earth and Google Maps, gravitating towards the product side rather than the software engineering side (JC still considers himself a recovering software engineer, working on little projects here and there). This is just a snippet of how JC went from the software engineering side to CEO of an AI company.

Coming Up with The Idea

In 2006, JC and his brother co-founded a company called T-Sciences where they both had the opportunity to work very closely with Google Earth, investing heavily in mapping. Through this partnership, T- Sciences had a picture of the world in 3D something that no one has had before. This was the time when Google Maps, directions and Street View (to name a few) blew up in 2006. Resulting from this new, powerful technology, everything from Amazon deliveries, DoorDash, Uber, and so many things we use in our personal lives were in the hands of consumers. However, there wasn’t a way that this geospatial location-based information had fundamentally changed the enterprise businesses operated.

When the business at T- Sciences was in the process of exiting, JC made the decision to join Digital Global, which at the time was still the world’s largest satellite imaging company. The product that JC was leading had a product for a platform that was applying AI and machine learning and big data analytics to 175 petabytes of imagery going back 17 years over the surface of the planet. This is obviously a lot of the content that you see in Google Earth, Google Maps, Apple Maps, etc. The goal, however, was not to show people the images, but instead: ‘How do we get information out of these images? How do we use machine learning? How can you extract interesting things like where roads are, where populations are?’ To JC, these questions were the step in the right direction, the step towards having something transformative for an enterprise. This was more of a platform style play where people would come in and write their own algorithms or create their own AI models and then run it in this environment on imagery.

Interestingly enough, one of the companies using JC’s product was a fortune 500 company called American Family Insurance. After using the product, American Family Insurance asked JC to meet with them, American Family Insurance had used the product JC created on the satellite imaging side to take aerial imagery and ground level imagery of all of their residential properties. They created the capability to fetch all of these images and run them through deep learning models to develop structured data, similar to what you would get if you sent an inspector out to the property about how many stories it is, the perimeter of the building, what type of roof it is, the material, condition, etc. This was done all through machine learning from the latest images available in the property. JC was beyond thrilled and excited. Here was a solution to do what he had been passionately longing to do for a long time, at least since he began working hand in hand with Google.

Testing the Product & Validating the Market

American Family had created a unique set of technology that they wanted to expand to other insurers around the world. At the end of the day, however, they are a property analytics company that’s applied to: insurance, the residential property, and casualty insurance market. If JC is going to value the property, not insure it, many of those things that make it insurable also contribute to its value. For example: condition, number of story sizes, etc. can help determine how much it’s really worth in the market. With that being said, American Family had new technology spinning out, which led to the founding of Arturo.

Over a period of about seven months, Arturo went to market and listened to the customers, engaging with companies like Hippo Insurance and their longest standing customers. While the IP agreement and negotiation was being figured out by big lawyers, Arturo was really understanding how to take a product that sat inside American Family and use it as more of a research capability that could work across a number of different companies. Ultimately, Arturo got spun out from American Family where they now serve customers both in the US and internationally.

Using their new technology, Arturo’s focus has been: ‘How do we price a property really accurately while really quickly providing a fantastic experience for their customers?’ The first part was to take the technology that was designed to serve one customer at American Family and make it scalable so Arturo could serve dozens or hundreds of customers in time, which came more from the software engineering side, making the platform and the infrastructure more robust to handle more inbound API requests.

Arturo also recently announced that they processed, with their client Suncorp, nearly 9 million properties – every residential property in Australia – within two days (48 hours). One year ago, this would have taken almost a year and a half to build out a data set similar, whereas now it is accomplished in two days.

Technology advances because of the effort put into Arturo’s engineering team. They are also figuring out from a product market fit perspective where the ability lies to suckin all images of a property while understanding what the property condition is. Arturo is focused on really discovering how to use this technology to scale from a software engineering perspective to serve all customers at once.

Challenges Encountered

There are so many perceptions that AI and Machine Learning are going to replace humans, which creates a fear amongst the masses that people are going to lose their jobs to robots. This, however, is not true. In fact, JC finds this ironic because at the end of the day, building startups are all about people. They’re about the importance of people and the relationships that they have, whether those people are your customers or your teammates. And no matter what you’re using, Machine Learning or AI, it’s still always about the people. A challenge is teaching people and companies the value of Artificial Intelligence.

As a first time CEO, JC has made mistakes, same with the members of his team. But the growth lies in how you come together as a team in each and every moment. For example, managing COVID has been a massive challenge, yet having a sense of togetherness has helped the team thrive. This is one of the biggest challenges JC has encountered. Hiring people during COVID where no one has ever been in the same room or met face to face. This can be challenging. However, as stated, it’s about building relationships with customers and team members while communicating the value of the product. It starts by recognizing and acknowledging the challenges in order to put time and effort behind them. JC values a great work environment and customer success team, one that is fully present to work with customers while making sure they’re successful using Arturo’s technology.

One of the things that was realized early was the ability for a customer to trust the outputs from machine learning, to know when they should trust them and when they shouldn’t trust machine learning. To JC, this is really important. Confidence framework was something Arturo invested heavily in.

Future Endeavors

Arturo’s future endeavors include growing beyond just insurance. The better the company understands physical properties in the world, the more other industries can benefit from their product. Now, Arturo has the ability to look at a home or property and identify any issues, this is before someone comes in to give a full inspection. This ability a huge use case Arturo sees itself moving into in the future. As Arturo receives its series B of funding, they’re able to address these markets more effectively. JC also thinks drones will be in our near future, collecting information about properties.

Luckily, Arturo is having fantastic success, keeping busy with customer demand and interest. With a focus on customer success, not just revenue, Arturo makes sure each customer gets exactly what they paid for, a continual focus into the future.

https://www.arturo.ai/

Introduction to Guest

Mike is the Chief Digital Transformation Officer at Khoros, a digital customer care service with AI-powered workflows and operational analytics. An experienced executive in contact center and CRM, Mike worked for MCI for 14 years leading contact centers and technology and has been helping big brands interact with customers for over 30 years. He was also recruited by Convergys as a President and Officer to create a new technology division and lead key acquisitions. Mike received his MBA from St. Edward’s University.

Coming Up with The Idea

Mike’s journey started in a unique way, in sales and marketing. In 1985, Mike went to work for MCI for 14 years and had an amazing journey. Mike helped grow MCI from a little company to an $8 billion dollar company in revenue on the consumer side. He was in charge of all the call centers and call center technologies. There were 26,000 people at MCI that were managing talking to customers every single day. Mike learned about AC, DS and IVR, call centers and workflow, and routing (to name a few). However, after 14 years, Mike sold his stock at MCI and started another company, which later was sold to Siebel.

Over time and into 2009/2010, Mike was working with big brands at the beginning stages of social media. These big brands were angry because people would go on Twitter and say things out of context, or statements that were not true. They would go on Facebook and post random things that started to destroy the brand’s marketing. The brands simply didn’t know what to do or how to handle the situation. At this moment in time, a light went off in Mike’s mind. He was going to build a company that had the rigor of a call center for customer communications in social media. In 2011 and in 2012, Mike and his team built a company called Social Dynamics, which was all about asynchronous messaging. All digital channels could come in and the brands could manage their social media platforms. They realized people weren’t trying to be mean, they were actually trying to engage and get brands to listen.

Social Dynamics was soon sold to Lithium and Lithium was later sold to Vista, which all live under the umbrella of Khoros now. Khoros have care, marketing, and community all living and working hand in hand together. Mike and his team began bringing digital customer engagement capabilities, through this new platform, forward in the world. This all wraps up the last 30 years of what Mike has been working on – building software to help brands engage with their customers.

Platform-based Approach

Khoros is an aggregator of all the channels. When they started, Khoros worked primarily with Twitter. Later came Facebook, Instagram, WhatsApp, and WeChat, as well as other messaging channels. While Khoros started out in social media, Khoros had begun primarily focusing on messaging channels, which is still growing rapidly. Khoros is able to think of a person at a brand, able to see every single channel and see the engagements – what people are saying on Twitter, Facebook, etc. – and then Khoros helps brands listen to their customers’ content, understand their needs and who they are. From there, they can prioritize converting these potential customers into clients who buy.

The social platforms that Khoros’ customers use depends on their particular industry. For example, Yelp and Google Maps are used in the restaurant industry. Facebook used to be big. Twitter is big in the US but not other parts of the world. WhatsApp is used worldwide, but not so much in the US. Basically, Khoros tells brands not to worry about social channels, for they have all channels built into their capability. It’s important not to worry about which channels to use, but instead focus on the people’s needs using a particular channel.

Validating The Market

Currently, Khoros is hosting all of the content and tracking all the engagements for brands. Khoros manages all channels, while keeping track of the brand’s history. They build tags around conversations, creating different profiles/personas for people who are interacting. From there, Khoros creates marketing content around that specific persona of that person. So, every time someone engages, the brands become smarter about their audience. This is then leveraged through Artificial Intelligence to make intelligent decisions based on that persona.

Khoros’ AI and technology helps narrow down what a customer really wants. To understand the customer, you must treat them like a person. This can all be done using AI. First tags are created, then labels, and then the labels are used to create content that is now leveraged for marketing, for example.

Working with Artificial Intelligence

Khoros has been spending a lot of time and energy in the AI department. The first thing they’re doing is aggregating all of the content. AI is used for two parts. Part one: what in the world of content can convert this potential customer? You can use Artificial Intelligence to understand what this person is asking for. You can see how they’ve engaged with content you’ve presented to them in the past. AI helps make these particular decisions, what content to present customers moving forward. The second part: using AI and/or bots to get an answer to a question from a potential customer. A lot of time people do not want to talk to someone, they just want an answer to their question (in messenger, for example) or they want to create content and share it. These are all ways in which Khoros uses Artificial Intelligence.

Advice for Aspiring Entrepreneurs

Find one thing and do it well. Be great at one thing – really strive to be the best. Stay true to your passions and your dreams. You will achieve your goals, but you have to stay on it. Sometimes the road is long but keep at it. Start small, find a problem and see how you can fix it. If you’re interested in AI, find out a meaningful way to plug it into the ecosystem.

Working with Industry Leaders

Let yourself evolve. Let your customers be in charge and listen. Embrace change management, new technologies, Artificial Intelligence, machine learning, cloud-based computing, etc. If you open up to new channels of communication, aside from using 800 numbers, the customers will go to these new channels. The customers understand social media channels. Embrace them.

https://khoros.com/platform/care

Introduction to Guest

Sam Guilaumé (Geelohme) – is the CEO at Aryballe, which combines biochemistry, advanced optics and machine learning to mimic the human sense of smell. Prior to Aryballe, Sam was the co-founder and CEO of Movea, growing the company to the leading motion sensing technology and eventual acquisition by InvenSense, the leader in inertial sensors for consumer electronic applications, Sam brings over 20 years’ experience in the microelectronics industry to the Aryballe team.

Coming up with the Idea

The call to action and thought behind Aryballe came from a good friend named Tristan. Tristan and Sam have been friends for many years and were running their own companies years ago when they first met. Later on, Tristan sold his company and was looking for his next big adventure. Tristan with a biological engineering background discovered a GPS technology used in the pharmaceutical industry – which led Sam and Tristan to wonder – what if this technology could be used for air instead of liquid? Tristan began to use the technology to make sensors, and soon brought Sam into the mix. This was seven years ago, and also when Aryballe was born. Aryballe is a mix of semiconductor sensor expertise with significant biological content.
Aryballe means a small container/bottle in ancient Greek to carry perfumes and sweets.

Call to Action

The idea to start Aryballe came from mimicking the principle of human sensing. The sensors in our nose are made of organic chemistry, which is pretty innovative when it comes to semiconductor, because organic chemistry and semiconductor do not bind together very well. When Sam & Tristan figured out that they could bind or graft organic chemistry onto silicone and make a sensor that duplicates the human sense of smell, it was all go from there. Aryballe has been able to duplicate all of the five senses, but the sense of smell is most tricky. Over the last seven years, Aryballe has been using an organic chemistry method of binding. Today the company has about 50+ employees, with offices in France, New York, and Seoul, South Korea.

Testing the Product & Validating the Market

During the first few years, the company has been technology focused, working on securing reliable technology at reasonable costs. This process took about five years, with the last two years investing in technology that keeps costs low and volumes high. This particular technology is only the size of a nail, comprising of a very small dimension.

Aryballe has approached their business opportunities in two ways: one is to duplicate another sense. The second being as humans, we collect and fuse together information, touch with temperature, vision with smell. With the idea that sense of smell will be as close as we can get to feeling and experiencing emotion. Aryballe’s game changing technology, we now have a system that can capture and eventually transmit and convey emotion. At this moment, Aryballe is in the transition of starting their industrial process. They are investing in a manufacturing line

Key Challenges

The lack of standardization and the lack of unity on how something smells has been the biggest challenge over the last seven years. You cannot describe smell with any kind of formula. So, unless you have the right technology and Artificial Intelligence, this feat is not possible. However, with technology and machine learning drastically improving over the years, Aryballe has been able to move forward at an accelerated rate.

The other challenge here is to graft these fragments of proteins onto silicon to make it efficient. The silicone is important because it keeps the costs low and the volume high. This is a chemistry-based challenge. The next challenge is to make the silicone effective, which requires a person with a specific expertise. Not to mention adding the need of electronics and machine learning engines. Aryballe’s challenges in developing the right product is a multidisciplinary feat. The next step wil lbe duplicating the preparation of a specific sample. Their are many layers that are crucial in production.

Aryballe employees’ chemists, optical engineers, electronic engineers, and mechanical engineers. Oh, and also people with a good sense of smell.

Future Endeavors

After Aryballe has proven that they can make their sensor product that is for mass market and compatible with many applications, the ultimate goals is to have a sensor that helps you in your daily life. Our health is criticial, and there has been a higher focus in living a more sustaible life. With Aryballe’s technology the hope to easily and effectively be able to make an assessment on one’s health condition.

Advice for Aspiring Entrepreneurs

Start with finding investors who are risk adverse, who can help you mature and become more appealing to industrial players. Invest in both strategic and financial investors, this concoction gives your company a proper balance. Also make sure you are very passionate about your product and surrounded by the right people. Know how to spot the right team players for your business.

Advice for Industry Leaders

Where do you see yourself five to 10 years from now? How do you want to contribute to the world? Where do you see your product/technology making an impact? These questions are not easy, but important. You must have a long-term goal and faith in what you are accomplishing. Faith in not just your future, but your contribution to the future as well.

https://aryballe.com/

Introduction to Guest

Mark Bainbridge is one of the Founders of Dragonfly AI, a predictive visual analytics platform that uses cutting edge neuroscience to accurately predict how the design of any content or experience influences what your audience sees first, across any channel.

Mark is also the CEO and Founder of Bainbridge Cooper Associates, a marketing and commercial growth consultancy, and Co-Founder of ProMarketer, an AI suite for the marketing and communications industry. Mark has over 30 years’ experience as an award-winning senior leader and CMO across a broad range of industries, from the British Army to Banking. The common denominator is that he helps solve reputational challenges. Mark is from the UK.

The Call to Action

Mark has worked for over 30 years with very interesting brand propositions. However, what Mark realized was businesses were slow when it came to change and adapt to new ways of thinking and doing.

In 2010 Mark was amazed, to say the least, how creative tech was emerging and saw an opportunity to get out there and witness a whole new trace of innovation coming to the market.
Mark left his secure job and came out in the world with a fresh approach to take on this emerging trend in creative tech.

Mark connected with a good friend who ran a company called Hack Masters, and soon joined the Hack Master’s network. Mark spent three years going around the world, looking at the markets and how they’re operating, hacking, and creating new business solutions based on forecast predictions. Really thinking about what the world is going to look like in 30 years. What will change? What will relationships with consumers be like? Mark was keen to produce something that held genuine meaning. With many people now looking at Artificial Intelligence Solutions, Mark was on an amazing journey of discovery. After a three-week expedition to the Algonquin National Reserve to clear his mind, it was then where he mustered up the courage and confidence to 100% throw himself into his new business, Dragonfly AI, and really go for it.

Coming Up with The Idea

Dragonfly AI was born out of a piece of academic research, looping at home to give robotic devices human visual interpretation skills so they could operate in their context, effectively, using human decision making. Mark kickstarted a piece of research, which lasted about four or five years, which helped him create the mathematical formula that the brain uses in a biological state. The research was presented at Queen Mary University London in 2014/2015 where Mark showcased the math formula that the brain uses to identify and prioritize what it sees in zero moment of truth. From there, they decided to bake it into a piece of software, turning the idea into an iOS application. You could now literally ingest anything you wanted to look at, giving you an instant view of what the human brain is seeing in sequence. A very powerful concept to say the least. You could use this technology in marketing, helping brands identify what their consumers see first in any channel and in any content environment.

Dragonfly AI lived with the iOS device for quite a long time. Mark reiterates how important it is to understand your market, for a lot of the early features they started developing were based on input they were receiving from their client partners.

Testing the Product & Validating the Market

Dragonfly AI began running a number of propositions at the same time. They had a bit of a false start. The first start that did not work well was Texels, a platform designed to try and identify all of the emerging technologies out there. Allowing businesses effectively to engage with propositions that they might not stumble across, if they were reaching under normal search trends. The challenge for a lot of innovation companies, AI, and technology-based solutions is they had identified a solution to a problem which they understood intrinsically and had solved it. The problem is, they had to sell it into a market, which is sometimes culturally and intellectually in a different place to where they are. But the reality was, their market – these businesses – take time to adapt and adopt new technologies. Originally Mark thought this idea was fantastic, but he was a bit ahead of himself.

It was in this moment, too, where Mark realized building the right team, one he could genuinely rely and trust, was important. Mark decided to give it another go. You’re going to have some good times and really bad times, and you really must believe in the product you’re putting out in the market. You must create a solution but ask the market whether it really needs it. The key is collaborating with the markets. Finding great people ties into this. By having at least five people on board, together they’ll feel more invested in what you’ve developed, while helping you understand the market you’re selling into.

Target Market

Dragonfly AI initially started out not targeting a specific market, but wound-up gravitating towards shopping marketing, and currently, ecommerce. Dragonfly AI can analyze how different assets are performing. It is perfect for pre-testing content before distribution. Using the software, you can see a heat map and look at a variety of different measures. Mark can hit a button and it will immediately show how the products are performing online. If he is a Brand Manager for Southern Comfort, for example, and his product is scoring a 51, it is not so good. It is hardly getting recognized. However, JIRA, a fine malt whiskey from Scotland, is scoring at a 93. So, this product is clearly grabbing attention. The user can deconstruct their page and drive by looking at all of the components, until they’ve got a complete understanding of how that page is functioning. Dragonfly AI presents the gift and ability to see the optimization of digital assets live.

Mark and his team are starting to look at things like sports sponsorships, branding, live events, etc. They’re working more and more with agencies to accelerate and optimize the brand partnerships.

Dragonfly AI is now getting more traction across both the brand and retail space. They have had new challenges to face, but their goal is to be as helpful as possible to clients through help, support, and solving their problems. It’s all about your customers. Mark wants to keep creating value while being a part of solutions for brands and companies.

Advice for Aspiring Entrepreneurs

Choose your partners well, because those are the people you will be working with for at least three to five years. Your business needs to be based on trust, adaptability, and resiliency. Be courageous when starting your entrepreneurial journey. Create a business that people believe in while trying to make your company a better organization.

Never give up. Build rapport with your clients and earn their trust. Hire people who have the strength to make decisions but will also ask for support when needed. Understand what/who you’re serving in the market. At the end of the day, create a positive, harmonious environment where you can get things done.

https://dragonflyai.co/

Introduction to Guest

Andrew is the Co-Founder and Chief Technology Officer at Qomplx, where he leads the development of next-generation operational risk management and situational awareness tools and information, sharing in multi-spatial, multi-temporal distributed systems. He is a thought leader in cyber and data management with specific expertise in large-scale heterogeneous network design, deep-web data extraction, and data theory.

Andrew was formerly a U.S. Air Force Senior Cyberspace Operations Officer who led enterprise network modernization and design efforts for the Air Force and large Dept. of Defense initiatives.

Andrew received his Bachelor of Science and Computer Science degree from the United States Air Force Academy and holds his PhD in Computer Science from the University of Oxford.

The Call to Action & Coming Up with The Idea

After a big opportunity to join the US Air Force and serve his country, Andrew fell in love with computers, cybersecurity, and machine learning. Upon leaving the Air Force, Andrew took an opportunity to go to graduate school at Oxford to learn more about computer science and study data extraction. It was at Oxford where he really dove into algorithm development, computational complexity, and other AI aspects. Andrew learned how to think about expert systems and how things can be automated to simulate human users, while at the same time learning about how we analyze heterogeneous data while dealing with inconsistencies and ambiguities.

During Andrew’s time at Oxford, he met his best friend and co-founder, Jason Crabtree, an incredible technologist, leader, and Artificial Intelligence enthusiast. Both Andrew and Jason knew how transformative AI is for our society yet were concerned as to why these technologies were not more accessible. Both military guys with leadership positions in cyber communities, Jason and Andrew decided to take on the challenging task of building a data driven decision system at scale. However, given the challenges working in this space in the US military, Andrew and Jason began talking more and more with people in the industry instead. Together, they made the decision that in order to have an impact on the world in a way they envisioned, it was not going to happen in uniform. Stepping out on their own was the best option. It was at this time where Qomplx was born.

Validating the Product

From Andrew’s time in the Air Force working on large projects, he really appreciated the central role that identity plays in cybersecurity. So, when it came to starting a company, Andrew knew they had to understand the cybersecurity market better. He ultimately made the decision that as much as the data factory is what the company is all about, it’s not just about the technology, it’s about productizing a lot of disparate technologies and making them interoperable with one another. Making them accessible so that customers can retain value and use their product with their data. It was a decision early on that in order to bring something back to market, they had to deeply understand the market, and it was better to build experiences on top of a platform as a means of understanding what the platform really needs to consist of. Given Andrew and Jason’s expertise in risk management and security, the first Qomplx products were related to cybersecurity and insurance, helping reinsurers reason about their underwriting risks. Automating underwriting so that underwriters have a better understanding of the risk accumulation of a new policy against their larger portfolio.

In cybersecurity, Qomplx built a product very centered around identity, using Active Directory data while looking at the underlying protocols by which authentication happens. From here, Andrew can help better assure that people are who they say they are, for example. This in itself is a very good representation of use cases for the larger vision that they are trying to promote. Cyber, as Andrew explains, is like the Wild West. There are sensors in different places all over the enterprise. Those sensors each have partial and very incomplete views of information. At times, they can be very inconsistent, even with each other.

Qomplx really began by finding market traction first, then built small products based on the market traction, products that were generalizable to greater domains.

Validating the Market

Qomplx took the approach of building the more prescriptive experiences first, with Q-Cyber being the first product that went into the market. Qomplx also had the best networking, too. Andrew was able to talk extensively with technology leaders in a plethora of firms that helped him understand the real need for potential customers. In fact, Andrew did not hire anyone specifically in sales until later on.

Cybersecurity: Finance, Insurance, Risk Management

Qomplx looked at the infrastructure they were building and wanted to showcase products that could gain market traction, with Finance, Insurance, and Risk Management the top three focuses. Andrew thought, if you do something three times, perhaps you can convince the market that it’s generalizable. Qomplx is not looking to build more verticals but proving that their company is generalizable. After this is completed, the right components are built underneath, allowing other businesses to build on top of the Qomplx platform. This is the ultimate direction for the company. Qomplx is not looking to be a product designer, but a technology provider for many different domains.

Pivoting Through COVID-19

Running a lean, agile product development shop, Qomplx has been able to pivot and gain traction during COVID, with the team working more closely together, productivity is enhanced. Qomplx is making sure their customers are taken care of, all while providing them the right value against their priorities. Andrew is pivoting through the challenges, while holding the reputation high that Qomplx is a responsible partner, even though the massive changes we are globally experiencing.

Vision for the Future

The future of Qomplx consists of not building more domain specific products but building and exploring more of the underlying operating systems. Andrew reiterates that the operating system is a bit of a metaphor. It’s an abstraction layer for tools and resources where one can find insights from data and enable data driven decision making. Andrew hopes to democratize those capabilities more and more, making them even more accessible. For it’s not just as simple as having a cool piece of technology but building a really great user experience that is accessible and valuable. Andrew and the employees at Qomplx plan to continue this journey.

Advice for Entrepreneurs

For an entrepreneur embarking on an AI startup, it’s important to understand that R and D is going to have to be continuous. As the market and technology changes, the competition will change too. It’s important that R and D investments are made. It is critical for AI startups to find the right investors, thought leaders and people that believe in your vision, who can help partner and support you in the appropriate way. Artificial Intelligence startups are not something that happen overnight. There’s a lot beyond the product market fit, for product refinement and experimentation is necessary, especially given the nature of these technologies. Make sure to find the right investors aligned with your execution plan.

Advice for Industry Leaders

Andrew’s advice for industry leaders is to understand the limits of your products/business. All decision making is highly contextual. When you start on this journey make sure your first several hires are not data scientists, but data engineers or vendors like Qomplx, that can help you wrangle, curate, and understand your data on a deeper level. Technology is constantly changing and transforming, so make sure to not become too dependent one one way of doing. If you have an infrastructure and data supply chain feeding what you know, your future Artificial Intelligence aspirations are going to be in alignment with your business and work.

As a whole, we must work on leveraging energy more effectively and efficiently, while using data and analytics to do so.

https://www.Qomplx.com/

Introduction to Guest

Daeil Kim is the technical founder of AI.Reverie, a simulation platform that trains AI to understand the world and make it a better place. Formerly, Daeil was a data scientist at The New York Times, and his research/learnings have been published in several of the top machine learning conferences (NIPS, ICML, AISTATS). Daeil received his PhD in computer science from Brown University, focusing on the development of scalable machine learning algorithms.

AI.Reverie was also ranked by Forbes as one of the top 25 Machine Learning Companies to look out for in 2020.

The Call to Action

At first, Daeil was not quite a technical person. In fact, he studied literature in college. From there, he transitioned into neuroscience – researching schizophrenia and neuro psychiatry, which led Daeil to start thinking more about the brain and fundamental aspects of intelligence. Daeil’s innate curiosity led him to pursue a PhD in machine learning, a transition that happened over a period of five years. After completing his PhD, Daeil wanted to apply machine learning to areas that weren’t being applied to. One of those areas was journalism. Daeil started his career at The New York times as a data scientist, working on solving their problems for over two and a half years.

During Daeil’s time at The New York times, he really wanted to go back to the idea of simulation. And with his advisor being a computer vision professor, he was aware of the fundamental problems they were dealing with. At the time, the big problem was data, which was a laborious process of labeling data. If Daeil was to think about supervised learning, as a way to train the state-of-the-art vision algorithm, the process within itself was inefficient. Daeil was determined to find a better way.

Coming Up with The Idea

Daeil was aware of these particular problems early into his PhD, and synthetic data was not a new concept. People in the academic world were trying to solve this labeling problem in computer vision. Others in the industry were able to create prototypes in academia around trying to use synthetic data to improve the training of computer vision. However, no one actually tried to make it a production system, meaning, how can we solve a lot of the world’s problem with it? This was an obstacle that Daeil was determined to tackle. Even though Daeil loved working at The New York Times, he knew branching out on his own was the best way to go.

Daeil spent several months learning game engine programming and how to build virtual worlds. He also met his co-founder during his stint at The New York Times (they both led international expansion). As a former entrepreneur and CEO of Giga home, Daeil’s co-founder was thrilled about Daeil’s vision. He, like Daeil, also left The New York Times to pursue his dream. Together, the dynamic duo founded AI.Reverie.

Testing the Product & Validating the Market

Daeil always knew there are ways to scale world creation, so the first thing he did was focus on the idea of taking geospatial data and recreating the world from it. From ingesting OpenStreetMap, satellite images, terrain data, for example, he could fuel the process of creating a world on a one-to-one scale in a very large area.

An example is the famous video game Grand Theft Auto. GTA spent over $100 million just making a small part of the world in terms of art, design, etc. in the video game. Daeil imagined something similar but with the mission to solve actual real-world problems in agriculture, retail, and government. Daeil was also able to use machine learning to accelerate the process.

The earliest adapters in AI.Reverie was the government, which surprised Daeil. The government was the one most open to the idea of using synthetic data. At the time, very few people believed in synthetic data as an actual solution to a product. Daeil was just curious to see who was interested in using synthetic data to solve computer vision problems. AI.Reverie learned that there are some interesting government applications around safety and started to understand how they could create the right product market fit. However, the market was not too prepared for AI.Reverie, it was a little too soon, ahead of its time. Companies barely had computer vision teams and were barely trying to understand how to set up those teams. It took a lot of phone calls and figuring out where the opportunities lie for AI.Reverie, all while creating an infrastructure that allows the company to work at scale.

One of the most important problems to solve is how to make synthetic data batter? It’s about laying out the infrastructure to pipeline, while trying to create a business. Daeil mentions that at some point, it’s a matter of getting the right clients, the right partners to work with you, and winning those contracts. However, this was not an easy process. It’s hard for every entrepreneur to find the product market fit, and it’s especially hard when you’re building deep tech.

Using Gaming Techniques & Machine Learning to Develop Virtual Worlds

Generating virtual worlds is an algorithmic process, and there are many things you must create before generating virtual world. However, using gaming technology to develop virtual worlds is different than the typical ways to build video games. The engineering process is very different from what you have to do for computer vision. For example, in video games you’re often focused on narrative structures, on storytelling, on creating an experience that people come back with. There’s a beginning, middle, and end. For computer vision there’s no need for that. What you really want to figure out with synthetic data is how do you create diversity? How do you create as many variations as possible of the vehicle? How do you create as many variations as possible in the world itself and the structure of the world?
AI.Reverie is using machine learning to figure out the right engineering pipelines of using gaming technology moving forwarding.

Challenges Encountered

The two main challenges encountered are: finding the product market fit and teaching/educating people about synthetic data, which can be a difficult concept for one to wrap their head around. Daeil did not think it would be so hard to convince people that synthetic data was the way to go. The main reason this is difficult being there aren’t any examples of synthetic data, or how it could potentially work and be an asset. That evangelism is still quite challenging.

Future Endeavors

AI.Reverie’s goal with synthetic data in general is to solve the world’s problems in multiple verticals, creating technology that’s more horizontal. To continue to create solutions for problems in agriculture, consumer packaged goods, retail, government (to name a few). Computer vision is a main focus because Daeil is thinking in terms of the future of robotics and Artificial Intelligence. You have to solve the perception layer first. If you don’t solve vision, robotics becomes a lot more challenging, for example, if you’re trying to ask a robot to retrieve a beer from your fridge. The robot at least needs to know what the fridge looks like. This is a huge ambition for AI.Reverie.

Advice for Aspiring Entrepreneurs

On a personal level, really think about your convictions and your beliefs in what you are doing and trying to accomplish. Spend time meditating on this because at times you will doubt yourself. You will wonder if it’s all worth it. When you feel this way, always go back to your first principles of why you think your idea or company is important and envision how to make it work in order to keep going. There will be sacrifices to be made, too, including financial sacrifices. You are taking a big risk. At the end of the day, be kind to yourself, and also be kind to others. Really believe in yourself and your company.

Mindfulness Principles

Become aware of you who are as a person. Self-awareness is crucial. What normally ends up happening is your own emotional triggers can become reactive, which is something not many people understand. If you don’t look at your past and your own past traumas, you won’t have the clarity to make the right decisions. It’s key to learn about yourself and to be kind and honest with yourself. Always be willing to do the inner world. For if you’re not personally aware of your triggers, receiving criticism or things not working out can cause you to put yourself down. This can lead to a downward spiral of negativity and self-doubt.

It is also important to note that if you’re not making mistakes, you’re not (in some ways) progressing. It’s ok to make mistakes – learn from them and keep trying to progress. Again, self-awareness is key.

https://aireverie.com/