Morgan Slade talks about NLP considering security measures. He underlines the importance of data science in the competitive era setting them apart from the rest.
1. Tell us how you came to be the CEO at CloudQuant. How much of your typical day is involved in innovating tech for your customers?
After studying engineering and finance at MIT, I spent over two decades as a hedge fund portfolio manager for a number of world’s largest hedge funds and having led some of the largest high frequency trading desks in the world. During that experience, I became acutely aware of the difficulties of harnessing the power of alternative data and machine learning in the investment industry. We essentially created CloudQuant to accelerate the conversion of data into dollars.
2. What are the applications or rather opportunities you seek to have with your product?
MOST VENDORS DELIVER “RAW DATA” TO CLIENTS AND EXPECT THEM TO TAKE IT FROM THERE. TO VARYING DEGREES, DATA CONSUMERS ARE ONLY PARTLY EQUIPPED TO TURN THAT DATA INTO INFORMATION, UNDERSTANDING AND ULTIMATELY REVENUE.
We fill that gap, by providing easy consistent data access into any technology stack whether it be on-prem or on the cloud, we provide our SaaS analytics stack as needed to fill any gaps and we engage our data clients as data scientists, AI experts and educators to help them understand what datasets they need and what they need to know about them.
3. How did you define the vision of CloudQuant? How did you approach your first 100 days as the CEO at CloudQuant?
We originally developed our tech stack via a relationship with Anaconda to support a growing institutional trading business. After investing heavily in cutting-edge technologies for many years to efficiently analyse alternative data in our investment firm, we spun that tech stack into a standalone technology company in early 2020. The first 100 days of our business was spent rapidly building out our repository of over 8,000 datasets and signing up our first set of anchor clients. Our Liberator data fabric has been embraced as an obviously useful technology for organizations of from the largest all the way to start-ups. It became clear very quickly that for data suppliers we were solving a problem they were facing of being able to measure and demonstrate the value of their datasets. For buyers we became a trusted advisor with free access to our educational research as well as a provider of the technology they needed to avoid bearing any upfront data engineering or integration costs to get their hands on novel datasets. One Liberator API gets them access to a world of alternative data.
4. What are some of the unique lessons you have learnt from analysing your customer behaviour?
Customer engagement is the most critical ingredient to monetizing alternative data. If you don’t have an engaged customer, if you don’t provide the data in a format they need to efficiently analyze the data and if you can’t help them get up the learning curve on how to convert the dataset into revenue, you won’t add value for them and generate revenue for the data vendors.
5. What are some of the distinctive features of CloudQuant Data Liberator (TM)?
Liberator is the results of years of engineering work by CloudQuant to solve the problem of rapid ingress and dissemination of data in a format that is research-ready. Liberator is a schema-less, multiclock, distributed data API with critical cloud and “last-mile” delivery capabilities via REST, c#, c++, java, python, R and Excel. It doesn’t just deliver alternative data, it has a stack of entity mapping and AI tools for keying raw data from a primary vendor source so that it is immediately useful to researchers. As veteran AI researchers ourselves we understand what most clients will need before they can do productive analysis.
The design includes support for live streaming data and that feature will be enabled for clients early this year.
6. And how do you differentiate yourself from your competitors?
The short answer is we don’t see any competitors who aim to fill this technology gap yet. We stand out because of the flexibility and efficiency with which we can onboard datasets onto our data fabric. It can be easily integrated into any type of research workflow or application so that more time is spent on meaningful analysis and presentation. Whether you are a fundamental investor operating in spreadsheets, a business analyst at a company or a PhD in AI operating a machine learning pipeline, our tools make you more productive. As long as we keep removing barriers to using new datasets we will keep our edge. That is why our clients commonly request that we add datasets to the platform for them so they don’t have to bear the cost of doing that in what is typically a much more expensive and time-consuming process.
7. CloudQuant in partnership with Vectorspace recently announced the availability of novel datasets that reveal relationships between global equity products. Can you elaborate more on the same?
Vectorspce has developed a unique suite of NLP technologies to identify emerging themes that can be mapped to set of publicly traded security exposures. They are using AI to surface topics that traditionally required human intelligence to sift through and filter large amounts of news to identify common patterns. For VectorSpave we are delivering a measurement of the historical value of these patterns identified by the VestorSpace AI engines to help our data buyers quantify the added value of the dataset. Then we are delivering their data directly into the client research ecosystem via our Liberator data fabric.
8. What are some of the common pain points that your customers commonly approach you with?
The reality is that most of our target clients are understaffed in the data engineering department. They may or may not have data scientists, but most organizations struggle to budget appropriately for data ingress and data engineering. It is viewed as non-core and non-differentiating. It is hard to get executives excited about spending money on it. We agreed, in most cases it is non-core and non-differentiating. At a high level, it is obviously
more efficient for the cost of data ingress to be consolidated under a single service and to have that cost shared across the entire userbase. No firm will ever compete with those economics and they don’t have to.
9. What advice would you like to give to the upcoming tech start-ups?
There is an infinite amount of technology that can be built, software that can be developed and products that can be created. Your goal is to produce the one technology, software or product that someone who has money to pay you can’t do without. Until you have real customers, don’t build anything you don’t need.
Hire a team with great chemistry, you will spend a lot of time together and you don’t have time for distractions.
10. Can you give us a sneak peek into some of the upcoming product upgrades that your customers can look forward to?
We have spanned the gap of rapidly increasing, mapping and delivering research ready data. Some clients still need more help further in the research process. We have additional services we are baking into our data fabric that will expose compute, AI and other SaaS products via the same client interface. Most firms are not staffed to develop some of these products themselves, but have a desire to get in the game. We are moving past the early adopter phase, but the investment and time barrier to design and implement some of these processes is intimidating. We offer a more efficient route to achieve results with alternative data.
11. Which is the one tech breakthrough you will be on the lookout for in the upcoming year?
We are big fans of webassembly (https://webassembly.org/) and we think it can play a big role in liberating ? Big Data by delivering high performance visualization and analytics tools. CloudQuant plans to contribute meaningfully to the opensource webassembly community over the next few years.
12. What is the one leadership motto you live by?
“Do everything you ask of those you lead” – which is a variation of George Patton’s famous quote. A small tech start-up in its early stages is more like a family than a company. In our initial engineering team we brought together talent from all over the world. Our best ideas have come from teamwork and having a common technical language with which to communicate. If we weren’t all operating at roughly the same technical level, the innovation would never be as effective and you would have one or two contributors. It may be my role to lead the organization, but I have an equally important role in helping to solve technical challenges we face. That’s the only way I know is to get in the trenches with the team and sort it out. That’s when the magic happens.
For more such updates and perspectives around Digital Innovation, IoT, Data Infrastructure, AI & Cybersecurity, go to AI-Techpark.com.