Interview with Moshe Kranc, CTO at Ness Digital Engineering

What are the current trends you’re seeing in digital platform engineering?

First, digital transformation is picking up steam. No matter what vertical your company is in, one of your competitors has probably raised the bar by announcing some new customer-facing digital product that gives its users new capabilities. In media, this may be the ability for artists to track their royalties in any region in real time. In financial services, your customers now expect to sign up for a new service and be approved within minutes rather than weeks. In manufacturing, your competitor may have installed sensors in what was previously a “dumb” component, turning it into an Internet of Things device that can be monitored and serviced remotely. Digital transformation is accelerating relentlessly, and every enterprise must up their game.

Another important trend is recognizing how difficult – but necessary – it is to combine the front end of digital engineering, design thinking, with the back end, engineering rigor. Designers are from Venus and engineers are from Mars – they have a different mindset, they speak different languages, and they are not particularly used to working together. Many companies have attempted to merge the two by simply engaging a digital agency to create the design and then handing it over to engineers to build, expecting harmonious interaction. The result is often failure and painful delays – the designers design wireframe interfaces that the engineers cannot implement, so the engineers implement their own interfaces which may/may not be user friendly. The landscape is strewn with failed digital projects, as the industry learns how difficult it is to truly combine design and engineering. But, it’s possible and extremely productive when the two are united from the outset, if you have the right framework in place. Ness has invested significant energy in figuring this out, and we’re achieving innovative outcomes with our clients as a result.

How does Ness’ approach to digital engineering stand out?

We know how to build digital products that solve real business needs, that are fit-for-purpose, easy to use, reliable, robust and scalable. To achieve this, we created a development methodology called Ness Connected, which defines the stages of agile digital development, as well as the handovers and artifacts for each stage. We have also developed a set of tools, called NessSMART, that provides monitoring, transparency, and feedback on agile development and continuous integration processes.

Beyond simply implementing what customers specifically ask us to implement, Ness provides even greater value by proposing alternative ways to do things. We provide this thought leadership based on our experience across many customers and multiple business domains. It starts with digital thinking – we can propose new ideas for how to digitally transform business, based on our understanding of business and technology trends and domain-specific expertise.

Engineering excellence is also part of Ness’s collective DNA. For example, Ness has been recognized as one of the best in the industry at distributed agile development. This is hard to get right, and it has taken us years to develop this expertise.

Ness stands out as a company that has successfully integrated design thinking with engineering excellence. Many others have tried and failed. At Ness, digital design and engineering are integrated from the top on down, from sales to discovery to design and implementation, via our Ness Connected framework. This vastly improves the quality of end products while reducing time to market.

Lastly, Ness can provide customers with thought leadership in areas like digital transformation, big data analytics, cloud technologies and machine learning. We bring a lot of added value through knowledge exchange, both internally and externally, that serves as a source for inspiration and challenging current assumptions to find better solutions. For example, our Chief Technology Office Associates program, which enables our customers to derive value from engineers across Ness to solve hard problems, is unique in the industry.

What role does big data analytics play in digital engineering?

Ness has defined three pillars essential to success in building digital products: customer experience design, platform engineering, and big data analytics. To explain how these pillars work together, I like to use the analogy of a car. Customer experience design defines how the car looks and how the driver interacts with the car. Platform engineering defines how well the car is built. Big data analytics defines what’s under the hood – what kind of engine is delivering what horsepower?

To unpack the car analogy, think of the expectations digital customers have when it comes to data. They expect the data to be based on as much available information as possible, including social media; they expect the data to be up to date to the current second; they expect to see this data within a few seconds of making a request; and they expect the information they see to be personalized to their specific interests and behavior. These expectations typically cannot be met with traditional relational database infrastructure. They require a new generation of Big Data Analytics tools based on technologies like Hadoop, NOSQL and Spark.

What will be the important trends in big data in 2017?

There’s some good news. Expect the big data industry to mature and consolidate in 2017. Gone are the days when any startup pitch including the words “big data” got funded. Venture capitalists have already placed their bets and are at the stage where they identify and fund the few winners, while shutting down the losers.

Customers are also starting to “double down” on a few popular platforms, while avoiding products lacking a solid installed base. This consolidation is especially evident around Spark, which seems to have caught on as the de facto platform of choice, both for big data in motion and for big data at rest.

As the computational center coalesces, expect to see more focus and innovation on the “edges” of big data analytics: ingestion and visualization.

Ingestion has always been the dirty secret of big data projects. At least half the time spent in any project is on ingesting and cleansing the data before it can be analyzed. Tools are emerging that accelerate data ingestion, using machine learning and Natural Language Processing technologies. For example, Tamr leverages predictive algorithms and crowdsourcing from subject matter experts to automate much of the work of extracting and cleansing semi-structured data for downstream analysis.

On the visualization side, tools are emerging that empower non-technical business users to gain insights via visual interfaces that encourage exploration and enable “mashing up” data from disparate silos. A recent entry to this genre is Amazon’s OpenSight, a cloud-based service that enables business users to easily connect to Amazon-based data, perform advanced analysis, and create rich omnichannel visualizations. Expect more products like this to emerge in 2017.

On the down side, the talent crunch for big data engineers will only get worse in 2017. Sure, there are a lot of resumes out there that include the right big data buzzwords. But, anecdotal evidence tells me only about 1 in 60 of those candidates actually knows what they are talking about. You will have to wade through a lot of candidates to find a few good big data engineers. Even when you find them, they will be attracted to lots of competing offers, as more and more enterprises jump on the big data bandwagon. Engaging a reputable partner can help ensure you have access to the big data expertise you need behind your programs.

What other emerging technologies will have an impact on digital technology in 2017?

The intelligence behind digital technology comes from algorithms for machine learning and Natural Language Processing. In 2017, expect to see the cloud giants (Amazon, Microsoft, Google, IBM) keep on competing with one another by releasing software tools that make these algorithms accessible for use in any application. Does your application need to understand human speech? There’s a utility for that. Does your application need to find correlations in data using deep learning? There’s a utility for that. If you are a data scientist, you may be more comfortable with Google’s TensorFlow’s programmatic interface. If you are a business user trying to put together a solution for supervised learning, you may be more comfortable with Amazon’s automated solution. Whatever your background, you’ll find a toolkit suited to your needs.

In terms of cloud technology, 2017 will be the year that hybrid clouds go mainstream. As users begin to mix and match proprietary tools from different cloud providers, they must manage hybrid clouds, where data and services for a single application are provisioned from multiple on-premise and cloud-based sources. So, expect to see the emergence of tools that enable users to deploy, manage and monitor hybrid clouds.

I believe we will also see Virtual Reality (VR) and Enhanced Reality leave the realm of games and entertainment and enter the enterprise mainstream in 2017. For example, suppose you are a brick and mortar retailer whose biggest problem is shoppers who leave the store because no sales person is available to interact with them. VR provides an excellent solution in the form of a shopping assistant hologram placed in a kiosk at the store entrance, who engages the shopper in a short conversation and sets the shopper’s expectations as to when a human will be able to help them.

What are the risks to success in digital engineering projects? How does Ness help mitigate those risks?

When digital engineering projects fail, it is often because of internal divisions between the business side of the house and the technical side of the house. They speak different languages, and they often have spent years blaming each other for missed business opportunities and long-term projects that never paid off. Because of Ness’s experience in talking both languages, we sometimes find ourselves serving as an honest broker to set expectations on both sides.

Another risk factor is the disconnect that can occur between designers and engineers. The designer creates a wireframe, and throws it over the wall to the engineer, who finds that the wireframe cannot be implemented as specified using the existing UI tech stack. Ness overcomes this thanks to its Ness Connected framework, where engineers are brought into the program in the design phase to ensure that what is designed can be implemented, and designers produce working prototypes using the target UI tech stack rather than wireframes.

Finally, there is a risk in choosing the right big data plumbing to support the digital project. In the traditional database world, there are safe choices. Traditional relational databases are like a spoon – you can use it to solve a lot of problems, and it will do at least a decent job no matter what the domain. Unfortunately, in the big data world, there are no safe choices. Big data products are more like a potato peeler or an apple corer – they are far more single-purpose, far more tuned to solve a single use case, and quite useless for other use cases. So, if you do not understand the big data architecture needed to solve your use case, you may end up with something that does not work at all. Ness can help mitigate this risk, thanks to our experience with big data products in a number of digital transformation projects, and thanks to our envision proof of concept approach, where we test the chosen big data technologies in an environment that accurately represents your data usage patterns.