Mike Branch is the Vice President of Data & Analytics at Geotab and leads the charge for developing solutions that enable insight from the over 30 billion telematics records that Geotab processes on a daily basis. He was previously the CEO of Inovex Inc., a software development company he founded in 2003 with expertise in the healthcare and energy sectors

Interviewing a data company like Geotab about connected cars is like opening the inside of a mechanical watch to see how all the moving parts work. The sensory network that needs to be built in a city to enable smart cars require correlated effort from a large assembly of smart city specialists. The computer vision technology needed to give smart cars the sensory input to make driving decisions is a field of its own. The computation and processing of data from car sensors and GPS required to give connected cars contextual awareness of their environment is an entire market segment driven by companies like Geotab. Geotab is a private company that develops, manufactures and supplies GPS fleet management solutions. The company specializes in telematics and analytics-driven fleet optimization, and is quickly expanding into the growing fields of connected vehicles and smart cities.

Analytics by Design (ABD) correspondents, Usama Malik and Jeremy Fajardo, interviewed Mike Branch, VP of Analytics at Geotab, to explore the world of connected vehicles and smart cities in Geotab’s point of view. Branch shared his successes in helping to steer the company through near-term issues like scalability and data privacy while navigating Geotab across the difficult terrain of the connected vehicle industry. The ABD team also got an inside look at how data teams at Geotab are effectively organized and managed to build cutting-edge analytics products that achieve ambitious business objectives.

Managing the Ecosystem of the Dataverse

Geotab manages the entire data generation and analysis process since the company also develops the telemetry devices that relay data directly back to the Geotab system for analysis. Geotab offers data products across the entire fleet management data lifecycle, such that services and products can be used modularly or in combination by their clients. Even if OEMs develop their own telemetry devices or leverage other third party sensors, data can still enter Geotab’s ecosystem through open APIs for enhanced processing and analysis.

“We position ourselves as an open ecosystem. Our success to-date has been through the platform we’ve created that enables our partners and customers to build some truly incredible applications that scale exceptionally well.” Branch said.

Branch believes that “the actual device or dongle used to record telemetry data still has a place in the market since the average age of a vehicle in the US is around 11 years.” Geotab will still have that part of the market. “But there will be parts of the market where we don’t need that device anymore because we get the data directly from the vehicle. So, what we’ve done is developed channels where that data still comes in. Whether from our device, from competitor devices, or from the OEM, data still comes into our ecosystem which we have built up as a great equalizer. It’s about the APIs, it’s about how people interact with the data. It’s not just about the data anymore but everything we have built on top of it.”

The goal of an open ecosystem that taps into different aspects of the data-driven technology stack goes beyond revenue generation. The vision is to create an environment that enables open data sharing and processing which are critical for connected cars in a smart city environment. Closed data silos not only inhibit this kind of data pooling. The lack of data from a multitude of sources also impedes innovation in, for example, the development of smarter and more robust algorithms.

“We believe in a partner ecosystem approach,” Branch added. “With all the organizations we partner with, we are experts in scalable data collection. In the connected car, we don’t employ computer vision at Geotab, but we have partners with that expertise and we infuse them into the ecosystem. It won’t just be vehicle on-board diagnostics (OBD) data that’s going to be important for making decisions in a connected vehicle. It’s going to be a combination of all kinds of sensory forces coming together.”

Branch emphasized that Geotab is an engineering company at heart and that’s what the company wants to grow as. The development of such an ecosystem is what allowed Geotab to grow as a company to what it is today.

“We focus on engineering a rock-solid platform, and our incredible network of partners bring complete solutions to market for our end customers. This model works exceptionally well.”

Driverless Cars: Where the Technology is Now

Branch painted a realistic picture of where autonomous vehicle technology is right now and where it will be in the next few years.

“I don’t think the driverless technology is going to be as ubiquitously available as currently predicted. I see people using figures like X% of vehicles by 2025 will be fully autonomous. I think it’s going to happen more in regions of cities and within more controlled quadrants,” he said.

The key point to note about driverless car technology is how long the ‘tail’ of the technology is. The sensors and onboard computing constitute only one portion of the entire technology stack. Behind these frontend technologies, there exists a sophisticated analytics backend to optimize routes, avoid pedestrians, reduce speed around areas where harsh braking tend to cluster, and so on. The data-driven computing to optimize routes for driverless cars (and even non-driverless cars) alone presents countless development and monetization opportunities for businesses. That being said, the lack of unified data technology standards for extraction of vehicle data presents a big challenge, and is an area where Geotab believes it can serve as a normalization layer because they have already done this for years.

“Our focus right now is to figure out how we can help in building an ecosystem for connected cars. We are not going to be in control of the vehicles, that’s not our game plan,” Branch said.

“Regardless of when the world goes autonomous, Geotab will still have a huge role to play. We see ourselves as a great equalizer,” Branch commented. Geotab products are interoperable and will remain desirable even if there is little headway made towards a single standard. “Whether you are driving a Honda Civic or a Ford F-150, the Geotab system will consider a seat belt buckle / unbuckle event as either ‘on’ or ‘off’ despite the fact that different OEM protocols are running in the background.”

Smart Cities: Infrastructure for 21st Century Urban Management and Autonomous Vehicles

Geotab is also making headway in the Smart City domain by leveraging its existing expertise in telematics and fleet management optimization. One of the most common themes in smart cities is the deployment of sensor networks across the city such that city managers and urban planners have massive data sets at their disposal. These data sets could be used for building analytics models that will help monitor at aggregate and plan for urban development projects by providing valuable insights into current city dynamics. Geotab helps accomplish this with its array of sensors on fleets driving throughout a city that gather data across the municipality dynamically. This sensory data helps optimize everything from intersection control to traffic signals.

“We have a dedicated business division catered to government customers and smart city projects that our data team works very closely with.” Branch talked about a particular initiative in virtual pneumatic tubes that was presented in Google Cloud Next last year. “Traditionally, we help government customers with optimizing their fleet management. But now as a result of having thousands of vehicles driving through around their city, we can (in aggregate) help them understand and better plan transportation infrastructure. We have a virtual pneumatic tubes application that measures traffic activity without having to deploy large crews to place physical tubes on the roads. After monitoring and comparing with Oakville’s actual pneumatics tube data, we discovered that if we had 50 vehicles driving in the area over a short period of time, we would reach coverage within the 85th percentile. You don’t always need full coverage to get a representative idea for important metrics that otherwise are collected manually. Applications like this and applications like intersection insights are interesting because it helps us model how aggregate driving behaviour and physical intersections fundamentally behave.”

From an infrastructure planning perspective, this information can help identify problematic intersections that the city can put more focus on without the need for additional sensors deployed throughout the city.

Another use case that Branch shared is the deployment of Geotab sensors on municipality fleets to gather valuable data like cellular coverage areas, air quality and temperature in each city block. Deployment on municipal fleets is very beneficial because fleets cover a lot of different travel patterns and move over the same paths repeatedly, which is important to collect good and representative readings. This initiative unlocked a new and wide array of data sets that is previously unavailable (or unavailable in real time) to city managers.

“Outside of the air quality work we are doing with the Environmental Defense Fund (EDF), we are also working with companies on other projects in this area. Our device is essentially an IoT hub, you can connect any kind of sensor you want to it. Whether it’s an air quality sensor or a temperature sensor, it’s all part of the same ecosystem. This allows you to map out hyperlocal air quality which, according to studies done by the EDF, varies block to block. So, enabling granular understanding at the hyperlocal level is very important.”

“Essentially, we see this data as the building blocks to map out what’s happening in the real world,” Branch emphasized.

“We have sensors that tell us whether you are going through a rough patch on the road and alert us on whether there is gravel or a bumpy road ahead. The sensors can even pick up whether an ABS is off in a certain area and can alert other vehicles in the area about it.”

Geotab is also working with intelligent traffic systems companies like Miovision in Waterloo in technology that allow data communication between vehicles and intelligent transportation infrastructure.

“If I’m a freight vehicle coming to an intersection and the light is about to turn yellow, maybe I should extend green for another 2 seconds so I can let the vehicle through. This type of control is useful because stopping and reigniting the heavy duty vehicle incurs costs, wastes fuel, and leads to increased idling, and emissions in the city. We are currently exploring ways to optimize Transit Signal Priority (TSP) through initiatives with Miovision”

Scaling Data Solutions

There are, however, more near-term problems that Geotab must tackle as a data services company, one of which is scalability and high availability of their solutions. The maturing of cloud technology has made data storage increasingly cheap and more accessible for businesses worldwide. However, for data companies like Geotab, the problem isn’t just storage. It is also speed. When dealing with real time data that needs to be processed at breakneck speeds, the main bottleneck is latency — the time required for data to traverse the system in the cloud. Latency will become the key metric as more analytics applications begin processing their data streams in real time. Companies are spending immense amounts of resources to lower latency and develop capabilities to process data quicker. Data processing for connected vehicles will not be an exception.

Geotab foresaw the growing issue of scale and continuously looks for ways to strengthen their infrastructure to withstand increasing data traffic.

“I think the biggest powerplay we have at Geotab is how we can scale all of this. A lot of people can do this for a few thousand vehicles or a few thousand drones. When you start getting up to millions, you reach a whole new level of problems in latency and scalability. We employ a lot of Kubernetes pods to dynamically manage scale in production. Even if we have 2, 3 or 4 million vehicles, we can sustain the load with our powerful infrastructure,” Branch said.

Implementing a scalable infrastructure is only half of the equation, the other half is hiring data engineering talent to operate and evolve the data platform. Over the last few years, Branch has organized a Data Engineering department at Geotab to focus on getting data through the pipeline as quickly as possible. To address problems in latency and real data streaming, he believes “it’s all about the ingestion and the pipelining process” which his Data Engineering team strives to improve relentlessly.

The Art of Developing Data Solutions

How do data companies develop data-driven solutions? The answer may surprise many new entrants and outside observers: Geotab spent YEARS just exploring the data. This is an enormous investment for any company to expend. Although a slow process, a thorough understanding of the data is essential if a data company wants to provide value to its customers.

“In order to understand the value we can provide our customers, we had to intimately understand our data. We had to develop the acumen and the toolsets before we can develop these solutions. Insight that you provide to one customer might not be the same insight that you should provide to another because to them, it might be of little value.“ Branch explained.

The company’s strategy in product development epitomizes a balanced synergy between product managers, data scientists and the end user or client. After exploring the data at hand in great depth, the Data Solutions team is divided into Agile pods that work independently to develop solutions for a specific sector or area of focus, such as smart cities or fleet management. “Without conducting the first few years of exploration, it would not be possible to identify these key areas of focus,” Branch said. Each team is guided by a product manager and comprised of a pro services and a data enablement team to develop the solution, starting with a MVP (minimum viable product). MVPs are based on self-generated ideas or proposals from customers or partners. Potential product ideas are deliberately shared with customers in order to foster innovation. MVPs also allow experimentation of new products and solutions early on in order to mitigate future risk and sunk costs.

“If they have a concept or idea that leverages our ecosystem, that they are committed to taking to market, and has a massive opportunity to scale, we will work alongside them to develop a MVP at no cost. And if we feel there’s something there, we pass it to the solutions team to scale,” Branch said. “The tricky part is deciding which ones we’re going to spend time on and which one we won’t.”

Opportunities in Privacy and Security

Outside of the traditional product development teams, Geotab has also stood up a new governance and privacy team. This team’s role is to interface with the Geotab privacy committee to ensure that any data leaving the ecosystem meets privacy requirements. This team also instills governance within development teams to ensure regulatory compliance.

A problem most data companies must inevitably deal with in the aftermath of major data breach scandals and growing concerns on proper data management is, of course, data privacy. Geotab is no stranger to oversight from government bodies that regulate data usage, cyber security and AI ethics.

“Just because we release aggregate data doesn’t mean privacy is automatically protected. You also need to ensure that you have consent and mechanisms in place that promote ethical use of data. Therefore, we employ third party companies that run privacy analysis on our aggregate data and verify for anonymity.” Branch said.

In talking about data security, the problem doesn’t just lie with building foolproof security for your network and ensuring that no threats can breach your system. It’s also ensuring that the data sets you utilize are aggregated well enough such that any traces of information that could be used to identify individuals are erased. This is particularly important for a company like Geotab that regularly releases aggregate data sets publicly.

“I have seen companies who claim that they are providing anonymous data and can pinpoint a list of where all the harsh braking is happening in a city. But if they do it based on any individual vehicle, it can easily be turned around,” Branch argued. He illustrated his argument in an example, “If I know that every Monday morning someone is slamming their breaks in front of a particular Tim Hortons, if I wanted to, I could position myself there and I could figure out who it is. Just because it’s anonymous, it doesn’t mean you can’t trace confidential information, and in turn, violate privacy. We do a lot of work to aggregate the data in a privacy-protected way.”

On the flip side, regulations, growing oversight and scrutiny of data policies are creating unique business opportunities. Data privacy, aggregation, and ethical use of AI have become huge markets that new companies are starting to exploit. Branch thinks the privacy area will experience a lot of potential market growth.

“You see GDPR in the EU, CCPA in California and you can tell that more and more of these regions are concerned about privacy…and rightly so.” Branch said.

Firms like Privitar and Privacy Analytics are using technology to evaluate data privacy more rigorously. “We see Geotab as a leader in the privacy area too,” Branch commented. “By working with privacy experts and applying these principles into our data strategy and delivery mechanisms, we can innovate in a way that maintains the trust of our incredible customer network.”

About the author(s)

This article is written by Usama Malik, Content Writer at ABD. He is a Masters student at Queens University in the Management Analytics program.

The interview and publication process was assisted by Jeremy Fajardo.

Want to learn more about the outlook of connected vehicles and smart cities?

Learn more about Geotab here.

Follow Analytics By Design and stay tuned for similar conversations with industry experts and thought leaders.

Analytics By Design (ABD) Toronto

Driving Data Solutions for Connected Vehicles — Mike Branch was originally published in Analytics By Design on Medium, where people are continuing the conversation by highlighting and responding to this story.

Eric Bogart (VP of Advanced Analytics @ ACOSTA) offers his insights on where retail is (and isn’t) rated so highly as a new frontier for advanced analytics application.

On the western fringe of Toronto in Liberty Village, Mosaic’s offices are nestled within a larger reclaimed warehouse — the shared home of many other businesses. But small might not be an accurate description of the design and experiential marketing firm, having been acquired by Acosta Sales & Marketing in 2016.

Acosta is no stranger to acquisitions and is an established firm operating between CPG manufacturers and some of the biggest retailers and grocers in North America.

Like many firms who play a consultative role in their industry, Acosta has made recent investments in advanced analytics as a core competency of their enterprise. We sat down with Eric Bogart, VP of Advanced Analytics for an open conversation about his career, the future for AI in retail and the Swiss Army knife nature of his team.

The rise of “Data Science”

Like many analytics leaders today, Bogart can recall what the early days of “data science” felt like, reflecting on the sense of community that exists today contrasted with what it was like in the past.

“[Back then] analytics was a discipline of engineering. So, statisticians were still called statisticians. The idea that we [different industries] are all trying to solve different problems in the same ways with a common toolset and/or philosophy didn’t really exist yet.”

We ask him what’s changed and he lists a few technological leaps. The ease at which he recalls them underscores just how important these two factors have become: the rise of open source and accessible cloud computing.

“Originally, all of the computation power came from your local machine. So, often times your ambitions were greater than your ability to solve them. You ran into a lot of situations where you might have a solution to something but without any way to scale it. ‘As long as the work comes back to me, I can always provide you the output.’ But the idea of shovelling that raw data into a system to do the work for you was sort of unthinkable.”

‘Unthinkable’ would have been a common descriptor for many of the capabilities machine learning powers today. Whether or not the broader promise of AI will live up to the vision it espouses is an ongoing debate.

P&G and the proto-Analytics community

“People would joke: well, maybe I’m a data scientist.”

Bogart’s tenure with P&G overlaps with the early days of data science hype. He describes the beginnings of analytics centralization at P&G as a ‘mixing of existing personalities’.

“About a year into my work there, they decided to group analysts into a shared services group. So, this was marketing analysts plus supply chain folks plus retail analytics — all these different personalities now starting to share a single identity. This eventually turned into the embedded analytics professionals P&G has now.”

He suggested that P&G may not have created ‘data scientist’ roles immediately as they had established identities as analysts.

Operations research and the perfect coffee bean

Bogart mentioned an early use case in Operations Research at P&G that reminds us of the optimization problems studied in current analytics hybrid programs like the Queen’s MMA.

“I remember one of the first major success stories back in the 1980s was a blending model for coffee. The idea was fairly simple. Different bean types have different flavour attributes and a brand — like Folgers — has a specific flavour profile. So, the model would look up the daily price of individual beans and optimize the cost to deliver a certain profile.

This ended up becoming one of the first Operations Research (OR) applications at P&G. So, by the time I joined (almost 20 years later) the OR team was already a well established group having evolved from projects like this.”

At the time, this would’ve been a monumental achievement requiring coordination from multiple teams and substantial IT costs. Today, this optimization problem sounds like something that could be achieved in a single python notebook.

Scalable data science

“It feels cliche to say it but investing in data science right now is really about shortening the gap between insight and action. It’s building solutions to embed good decision making into processes.”

Bogart illustrates why that shortening is so crucial in retail.

“You can produce great analysis that says: we have some SKUs with too much inventory. But now you have to go the inventory planner and present that work. Maybe she doesn’t believe you and maybe she’s right, maybe you missed something. The issue isn’t about who’s right or wrong — it’s the delay because you weren’t on the same page somehow.”

Retail has many broken telephone-esque instances like this due to the siloed nature of how retailers operate. To Bogart, the solution is a question of system design and agile execution.

“You need to get everyone to agree to the premises of the analysis and the underlying philosophy behind what we’re doing. Once you have that alignment, then you can build the pipeline to support the MVP of the system that automates the result or insight you were originally digging for, and then — very importantly — iterate.”

This methodology is what Bogart’s team strives to implement in their own work. Considering the scale of what Acosta does with its store reps all over North America, strong consideration for scalability not only makes sense but is necessary for success.

The gamefied store rep

“We have thousands of reps calling on stores all over the country. It’s a workforce with a good mix of part-time and full-time workers. Their expertise is to find problems in the stores and fix them.”

Supporting this work force with traditional analytics outputs like reports and dashboards doesn’t really make sense. So what is the product that empowers an individual rep to address the right problems?

Bogart describes a system that generates a curated list of tasks that are informed by data and can be tackled by individual reps.

“Consider a rep that has a limited amount of time to spend in a given store. They might not be there every day, maybe it’s a weekly visit, maybe there are issues with different item categories. How do you help them optimize their time and address the right issues? How do you empower them with insights from our latest prescriptive analytics that will help them make better decisions quickly?”

Bogart never explicitly mentions incentive systems or details on how a system like this would be productionized (perhaps an app), but the underlying idea of decision automation / support for human labour is a compelling example of scalable analytics and automation with human oversight firmly in the centre.

“If I give a list of 100 things for a person to do everyday, that’s probably not efficient. But if we look at the average time to completion for every individual task, crossed with each rep’s own historical productivity at different times in different places, can we curate a list that’s more optimized but manageable? More importantly, can we drive a sense of satisfaction for own employees?”

The key idea:

“If a system can be built and leveraged by a business without constantly having to teach, inform or persuade. That’s a much more scalable way to influence behaviour than traditional analytics outputs.”

Where AI will be most effective in retail

We shared an article with Eric that’s made the rounds in the analytics community to get his take on it. He laughs, noting that he’d seen it earlier today — his boss had forwarded it to him.

“There’s a few things that drive those recommendations [from the article]. The reality in retail is that there’s a lot of effort that goes into repeated processes with a lot of rules-based decisions being made along the way.”

Bogart suggests that AI’s most valuable trait for retail is the flexibility it can create within existing systems and the potential to replace the current rigid, rules-based systems with AI-driven ones. A less sales-pitchy way of saying it is: analytically automating the simpler decision points to the best of your abilities (or within your acceptable error tolerance) can help to expose the parts of the process where human intervention is most valuable.

“All these things that are AI-friendly but are currently being done by humans is a big part of why the valuation of AI in retail is so high.”

The second key trait of AI that Bogart believes can be better exploited are feedback cycles.

“Ultimately, this is the higher promise of AI right now, right? The idea that any sort of model(s) that is put into production somewhere can always be tweaked, revisited or retrained provides some amount of scalability into the future.”

These are the bigger picture ideas. There is, of course, mention of the potential of robotics and vision systems that will certainly be disruptive in the future. Bogart acknowledges this as well.

“ […] these are the things outside of the more nuanced side of retail; things as simple as counting out-of-stocks, for example.”

The more nuanced side of retail that Bogart alludes to is the world of marketing, forecasting and branding. Broadly speaking, the consumer facing side of the industry.

“This side of the business deals with things like: what is the effect of adding this new item to a category given its own set of characteristics? Yes, there are models for this already that could be augmented by AI. But ultimately, this is the side of the business that is core to brand and identity. ‘Who are we and how do we attract and retain shoppers?’ This is not a question that can be directly answered by throwing AI at it.”

The search for in-store data

We then chatted at length about the challenge in capturing in-store data, but the sense we got after circling around cool use cases and potential solutions is that, there is really no clear and obvious winning approach yet. A shared sentiment that was echoed in another interview we had with Peter Cuthbert (Director of Customer Loyalty @ Loblaw — article coming soon).

“Lots of solutions right now rely on bluetooth or WiFi. There’s also video capture and proximity sensing, but these are things that don’t really address the who. It’s more about tracking the ‘blob’. How long do people stay in front of a display? What parts of the store get the most traffic?”

Bogart concedes that there is a lot of promise in solutions like these but the right convergence of methods hasn’t been found yet. He identifies one reason why finding concrete value is difficult.

“Scaling anything is hard. Anything that goes from a pilot to multi-store wide execution becomes magnitudes more difficult.”

If scaling is out of the question, proving value needs to be done with an MVP. But to Bogart, the tradeoff needs to come at the cost of trying to answer a harder question.

“Driving convergence between behavioural and sales analytics is one of the biggest frontiers. Mapping an individual shoppers journey through a store with some granular level of detail and linking it to the their final basket is huge. Being able to classify a shopper’s mission when they walk into a store, these are the harder, much more valuable problems to solve.”

If we consider a point-of-sale dataset, a small basket could be associated with two different consumers. One could be a simple stock up trip and the other could be a dissatisfied shopper who left without buying everything they wanted. The difference between these two outcomes is that one is a valid mission and the other isn’t. In-store data is the key to separating the threads here.

So when it comes to in-store data, it seems the endgame outcome is clear. The difficulty comes from not having a roadmap of the best intermediate solutions along the way. Slowly but surely, it seems like retailers are willing to try things out.

What retailers need to do to win online — frictionless

“Wal-Mart.com might not be the best online shopping experience but it’s not too dissimilar from what shopping at an actual Wal-Mart is like. It’s not overly premium online and sometimes the aisles can get a bit confusing, but online and in-person don’t feel like two different stores.”

Bogart is quick to point out that building a carbon copy of your in-person experience online is not the point.There are retailers with vastly different online and in-store experiences and are majorly successful.

Sephora, for example, adopted a showroom model wherein the retail experience serves a fundamentally different mission — the experience of testing and trying products. It continues to succeed because, while the retail experience isn’t primarily built to facilitate the final purchase, the overall experience of Sephora never gets in the way of it. In fact, reps in Sephora stores can easily facilitate online purchases and are encouraged to do so.

When assessing how to execute eCommerce, Bogart offers the following, “Retailers need to understand why that shopper is in their store and make sure to always support that mission. Then they have to figure out ‘What is the consumer’s appetite to shop online and whether competitors are offering similar experiences that are threatening?’

If the answer is ‘yes’, they need to ask themselves, ‘Does my brand extend itself in the same way?”

Lastly, Bogart makes a distinction about where AI is most effective in this domain.

“Advanced analytics can help with all the things that aren’t about the core identity or value proposition of the company. Once that is sorted, it falls on the right cross-functional teams to determine the right KPIs so that analytics can properly measure and inform you instead of mislead.”

About the author(s)

This article is written by Jeremy Fajardo, Content Writer at ABD. The interview and publication process was assisted by Mia Meng.

Want to learn more about the outlook of retail analytics?

Learn more about Acosta here.

Follow Analytics By Design and stay tuned for similar conversations with industry experts and thought leaders.

Analytics By Design (ABD) Toronto

The Promise of AI in Retail was originally published in Analytics By Design on Medium, where people are continuing the conversation by highlighting and responding to this story.

Douglas Heintzman is the Practice Lead for Innovation at the Burnie Group, a management and technology consulting firm in Toronto, Ontario. He is also chair of the selection committee for the NSERC Synergy awards for Innovation and a jury member of the Robot of the Year, the first international prize that rewards the best innovations in ethical artificial intelligence and robotics, beneficial for humans across 11 industries worldwide.

For over half a century, humankind has pondered the “singularity”, a doomsday scenario where an artificial superintelligence changes the fundamental nature of civilization in unpredictable and potentially dire ways. Almost 70 years ago, the pioneering computer scientist Alan Turing was asking the question “can machines think?” This question, and the consequences of its answer being “yes”, has not only lead to practical questions such as the need for a “kill switch” as discussed in AI control strategies, but has also been fodder for both philosophical debate and science fiction — often combined in tropes like Asimov’s 3 Laws of Robotics and Hollywood blockbusters like Blade Runner.

Fast forward to today, technological advances continue to fuel our imagination and challenge our preconceived notions of general intelligence and machine consciousness. We continue to speculate upon the potential of AI technology and the societal revolution that it will inevitably trigger. What happens when we invent an intelligence that is smarter than us? By definition, this machine will be smart enough to invent a machine smarter than it is, and in turn that machine will, again, invent a machine smarter than it is, and so on. How long will it take for an artificial intelligence to be so different from us that we can’t relate to it in the same way that an ant can’t really relate to a human? Is AI an existential threat or is it a huge boon to human productivity and welfare? And finally, what technological, business and socio-political problems should we think through if its advent is an inevitability?

To consider these questions, the ABD team recently interviewed Douglas Heintzman, a thought leader in disruptive technologies and AI. When we asked him about the singularity he was very thoughtful but ultimately dismissive, “There are so many steps between that kind of world from where we are today and we don’t even have many ideas about what those steps are. I suspect we are at least a century away from having to worry about Skynet.” In the meantime, Heintzman thinks there are plenty of things to worry about ranging from ethics to governance. Heintzman shared insights about the immediate challenges and opportunities that business and society are facing, and the need for a new social contract as we enter the Fourth Industrial Revolution. He believes that both our economic and political systems will need to substantially evolve in order to fully benefit from AI’s potential, and that the changes driven by AI will be upon us sooner than most people imagine.

Today’s AI Challenges

“Today AI is great at performing many tasks for which it is specifically trained. That’s great. It helps optimize air traffic, helps doctors diagnose cancer and allows us to ask a home appliances when the final season of Game of Thrones is coming out. In the real world, the problems we face are more compound, sophisticated and filled with all kinds of nuances”, Heintzman said.

“We are still a ways from truly generalized engines that we can submit compound problems to. In the meantime we need to work on the potential weaknesses, functional deficiencies, and the ethics questions those weaknesses imply.”

Today’s AI engines are specifically designed and trained by a selected data corpus to do specific things. Heintzman shared with us that “one the biggest problems that AI faces is the bias that may be hidden in training sets”. When an AI engine has been trained by a bias laden data set — and bias in this context may simply be an under sampling of a particular group — the AI engine will come up with the wrong answer or recommendation or take the wrong action. Creating high quality training data sets can be a tricky business. There is complexity and effort in transforming a massive dataset with fully labelled predictors and responses for use in AI. Following the notion of “garbage in, garbage out”, Heintzman is concerned that AI engines can learn the wrong types of lessons if the sample data used is not vetted to adequately contain patterns of interest, or is simply not large enough. Unless the data is heavily fine-tuned, sometimes massaged with synthetic data points and attributes, the engine’s success is hugely susceptible to any biases inherent in its training data. As Heintzman says, “the AI engine might design a substandard marketing plan, make a faulty medical diagnosis, or drive a car off a cliff.”

Echoing Heintzman’ concern on the risk of persistent bias, McKinsey’s Risk Insights argues that machine learning algorithms used to predict behavioural outcomes are also prone to bias. This is due to its heavy dependence on historic patterns, and the persistence of prejudices carried forward from criteria initially programmed by its human architects. Predictive models generated in this manner fail to recognize new patterns that are absent in historic data, and reinforce the same biases under the assumption that things will function more or less the same as before. For example, a social media recommendation engine that filters news based on existing user preferences will naturally encourage confirmation bias in readers. In turn, this will amplify stability bias in future recommendations.

Though it is no easy feat, it is possible to lessen data biases through not only rigorous data cleansing but also through advances in algorithmic design. Heintzman is especially excited about the potential of unsupervised neural network architectures, such as generative adversarial networks (GANs), that are promising to combat such biases. GANs are very clever algorithms that train each other in a double feedback loop. They create an adversarial learning environment for two independent neural networks, a discriminator and a generator. As Heintzman explained, “The generator tries to fool or prevail over a discriminator and the discriminator tries to detect when they are being fooled. A generator might, for example, be instructed to create a photo-like representation of a flower. The discriminator tries to detect the fraud. As these two networks iterate this process over and over again, the generator gets better and better at creating flower photos and the discriminator gets better and better at detecting frauds. These two engines can train each other very well, very quickly.” A notable example of a GAN is DeepMind’s AlphaZero, which generates self-play games (Go, Chess, Shogi) and in parallel trains networks, all without access to exhaustive data in opening and closing board positions.

Another significant challenge that Heintzman and other AI professionals are concerned about is known as the “black box problem”. Whereas a traditional computer program executes a defined logic tree which can be examined to understand why a certain decision was made, AI engines that learn from large training sets and from experience can be “black boxes”. It may not be completely clear why it made a certain decision or recommendation. As Heintzman pointed out, “This opacity of decision making introduces some significant challenges around the assignment of accountability”.

Ethics, Data Privacy and Rise of an AI Duopoly

The first reported fatal accident involving an autonomous Uber test car and a pedestrian in Arizona raises this question of accountability. Putting aside the question of whether a human driver could have avoided the collision by not watching TV at the time of the accident, and the reality that autonomous vehicles can see better, make faster reactions and are statistically safer than human drivers, the question of accountability needs to be addressed. Was the diver responsible because of inattentiveness? Is the software maker responsible for discarding the sensor reading as a false positive? Was the sensor maker responsible due to a lack of resolution? Is the car maker or car owner responsible for putting the car on the road in the first place? Heintzman argues that “the sensor recordings will likely help in assessing distribution of responsibility in situations like this, but there will be situations where we don’t understand why an AI made a certain decision. Because of AI’s potential in dramatically improving transportation safety in general, we as a society might simply have to accept some degree of unknowingness and embed that in our insurance and judicial systems.” Still, it is becoming evident to Heintzman and other like-minded technology and policy experts that there is a lot of room for growth on the governance and regulatory front, particularly in handling cross-disciplinary complications.

An even more pressing concern for many people is AI’s role in a “big brother” society.

“China is one of the largest investors in the development and deployment of AI”, Heintzman explained.

“They are using AI driven facial and behavioural recognition systems to monitor behaviours and actions of various minority populations. The data collected from this type of surveillance is used to quantify social reputation at an individual level. This directly dictates one’s access to certain levels of housing or education.” Heintzman and government leaders are getting increasingly concerned as the public under surveillance ponders, “Is this a world we want to live in?” Heintzman argues that the answer is actually “yes”, at least to some extent. “The convenience and value added to public safety, transportation infrastructure and medical outcomes for example, are very compelling and will have to be balanced against privacy concerns and potential misuse by governments and companies.” To their credit, governments are starting to tackle privacy issues more seriously but anything approaching a unified standard and an agreed upon approach have yet to merge.

After the European Commission introduced the General Data Protection Regulation (GDPR) to give data ownership back to consumers, policy leaders in America and China followed suit. According to the Council of Foreign Affairs, American policies, conventionally known for supporting big tech in a laissez-faire manner, are being re-examined by the nation’s top policy makers due to mounting public pressure related to a series of data breaches and privacy intrusions involving companies such as Facebook and Cambridge Analytics. China’s Personal Information Security Specification was put in effect in May 2018 and is somewhat similar to GDPR but was designed to be less cumbersome for businesses. With such divergent political and economic agendas, new policies are bound to emerge as world powers and tech giants learn to balance data monetization with privacy regulation. Heintzman argues that progress in this area is critical especially in light of the possibility that “the vast majority of AI created value and wealth will flow to two super states, China and the United States.”

Heintzman isn’t alone in his prediction of an AI driven economy dominated by two economic superpowers. Kai-Fu Lee, the founding president of Google China, also foresees the rise of an AI duopoly in the global political landscape where China and the United States become the dominant players. China is well beyond its historical dependency on IP theft to power its technical innovation. The world’s second largest economic superpower has gained tremendous strength in AI through massive investment in research and development. Moreover, China has by adopted a business model that is based on rapid iterative product design, and is informed by vast market data harvested from the world’s largest population. Lee predicts that China will eventually dominate the markets in South East Asia, Africa and partially South America, leaving United States with Europe, Australia, and North America. Heintzman postulates a possible implication from this situation where the economics of AI may lead to a superstate vs client state reconfiguration of the world order.

“We are already seeing international trade in data. In the not-so-distant future, data will be treated as a commodity. There is so much economic value will flow from AI driven decision making. It is entirely possible we will see a situation where two superstates will agree to fund comprehensive welfare systems of their client states in exchange for market access and access to data.”

How the future of the world’s economic and political landscape plays out, is as yet, uncertain, but what is certain, is that competitiveness between the United States and China will undoubtedly accelerate AI development.

Jobless Economy and A New Social Contract

“AI has been referred as a key part of the Fourth Industrial Revolution and with reason”, Heintzman said. “Vast computing power, digital and mobile connectivity and instant access to the world’s collective knowledge has caused technology to become deeply integrated into our economies, societies, and governments. Disruptive technologies such as AI, IoT, robotics, blockchain, quantum computing, and 3D printing are reinforcing each other and advancing at an unprecedented pace across almost all industries.”

This revolution has the potential to dramatically improve the quality of life but it will inevitably set off a chain reaction with significant consequences to labour markets worldwide. A key consequence, as identified by leaders of the World Economic Forum, is inequality. Heintzman concurred, “The economic benefits of AI will likely be unevenly distributed, leading to inequality.” Inequality has also been a feature of previous technology disruptions. However, Heintzman argues that this disruption will be different in both speed and magnitude. This will lead to unrest due to widespread fear that humans will be replaced by automation and machines, resulting in a “jobless economy”.

We are already seeing fear manifest around the world in anti-immigration and anti-globalization political movements. Heintzman noted that “Most participants in these movements fail to appreciate automation, the real driver of the disruption they are reacting to, and its impact.” He cautioned, “You need to consider the historical context. When we look back at previous industrial revolutions, the world order changed quite radically every time. Both capital and labour were liberated to do more productive and innovative things. For example, as automation and improved agriculture productivity reduced the labour intensity of farming, labour and capital were redeployed to higher value pursuits.” He also thinks the notion of fixed labour displacement is a “lump of labor fallacy”. Labour markets will ultimately organically redistribute itself and find a new equilibrium. If history is a guide, we will ultimately be more productive and wealthier as a collective. Heintzman argues that the speed and magnitude of this disruption “will make it especially hard for policy makers to navigate troubled waters, as unemployment, at least in the short to medium term, increases across large swaths of the population.”

The World Economic Forum pointed out that the fruits of automation will not be distributed equally around the world. It will especially disrupt developing economies. They can no longer monetize their once valuable manual labour force as corporations in developed economies leverage AI and other automation technologies to move production and services onshore. As the inequality in labor markets and economies manifest, the gap of the rich and poor will widen according to the 2018 World Inequality Report. This growing inequality breeds social distress and feeds the growth of extreme anti-authority parties. “This phenomenon imposes a global threat as these actors can express their displeasure not only through terrorist activities but also through cyber warfare”, Heintzman added.

How do we deal with some of these disruptive consequences of AI? Heintzman has some ideas.

“The Second Industrial Revolution, the war that followed it, along with the economic realignment of the 1920s and 1930s brought upon the need for a new social contract. In this contract, Western democracies introduced reforms in public works, social insurance programs, and universal education and healthcare. In a similar vein,, we are going to need a new social contract for the Fourth Industrial Revolution that puts forth corrective guidelines to offset inequality at large, and promotes constant retraining and lifelong education.”

The idea of a new social contract has many advocates including the Global Future Council on Technology, Values and Policy. It has preliminarily proposed two changes in the contract:

1. Establishment of a new Licence to Operate to enforce companies to take on social and environmental responsibilities;
2. Creation of national and global Citizens Wealth Funds to finance social services and infrastructure.

Progress in this area will be difficult. It will take some extraordinary leadership. As the World Economic forum made clear, there are many powerful forces, such as large corporations, that have advantageous stakes in the existing system and have considerable influence on policy makers. These forces will resist changes to the existing social contract and, in some situations, have already worked diligently to dismantle it. The new contract needs to recognize that in a world with AI and related technologies, “no man is an island”. The benefits of technology driven productivity cannot exclusively flow to a small elite. As Heintzman told us, “If it does, societies and our global economy will be destabilized. The cost of that destabilization, both in terms of human welfare and wealth, will be enormous.”

So what do governments need to do to prepare our society to be AI-ready and protect the revolution’s left-behinds? In Heintzman’s view, “We must get from this current era to the promise of the next. The future is inevitable. The road ahead will be bumpy. We are on the cusp of a very exciting future. We have choices ahead of us and it is up to us to make the sensible ones”. He hopes that policy visionaries and advisors, as opposed to term-based and interest-controlled policy makers, will emerge to get us over this transition period without ripping the fabric of society apart. And we should begin by designing and testing parts of a new social contract that will ideally address a majority of the problems to come.

We concluded our conversation with Douglas Heintzman where we began, talking about the impact that specialized and generalized AI will have on our lives. Heintzman believes that AI in many forms will pervade almost all aspect of our lives, however, true generalized AI is a ways off. He did concede and said,“We are starting to see global AI networks like SingularityNET emerge. SingularityNET use swarms of AI engines powered by a blockchain reputation engine to tackle more complex compound problems.” Due to the pervasiveness and the pace of development of AI , Heintzman believes that “we need to think through the many ethical issues including bias and accountability before it is too late.” He foresees heavier investments in the near future to better address the way machines should handle user/self-preservation priorities and ethical ambiguities.

The final thought he left us with us was:

“We need to engage in a formal and informed dialogue. The topic of the exciting future that we are heading towards, and its many implications on business and society, need to become a core part of our political and policy discourse. As many people as possible need to become literate in this area so that that discourse is productive and impactful.”

About the author(s)

This article is co-written by Kitty Chio, the Content Lead at ABD, and Michelle Liu, the President of ABD.

Want to learn more about the outlook of AI?

Watch Douglas’s keynote presentation on “Living in the Age of Disruption” at the ABD 2018 conference.

Learn more about AI here.

Follow Analytics By Design and stay tuned for similar conversations with industry experts and thought leaders.

Analytics By Design (ABD) Toronto

A New Social Contract and AI’s Most Imminent Problems with Douglas Heintzman was originally published in Analytics By Design on Medium, where people are continuing the conversation by highlighting and responding to this story.

Technology’s ubiquity is difficult to ignore these days and analytics — one of its key enablers — is an industry agnostic driver whose impact is probably best characterized by its explosive growth (and fair share of hype). Despite that ubiquity (or maybe because of it), trying to grasp the full scope of the latest trends, pitfalls and insights feels like an impossible task. In May last year, our team ran a conference to try and get a pulse on the industry from as many perspectives as we could fit into a single day.

Below are three key topics that arose and are especially relevant to practitioners at the enterprise level:

• Data Science Vendors
• Effective Visualization
• Quick-Wins and Technical Debt

Data science as a platform and service

In Toronto’s garment district, the AI firm Dessa occupies a refurbished warehouse where cofounder Stephen Piron’s grandmother used to work as a seamstress during the early days of the industrial revolution. It was a time where machines were being leveraged to augment the way humans worked and had a profound impact on society and the way we lived — echoing our current digital age with the applications of ML and AI.

The portfolio of Dessa and AI firms like it, showcase the breadth of possibilities analytics can have in multiple industries. At the ABD 2018 conference, Piron highlighted the blueprinting process his team follows and the “Dragon’s Den” style pitches they employ to filter ideas for AI applications internally. An engineer at heart, he noted the shortcomings of open-source out-of-the-box solutions and the subsequent pivot his team made towards building a software platform for enterprise AI (aptly named, Foundations).

The SaaS model is a common one for many AI firms as it provides an onboarding mechanism for enterprises to quick start AI projects with clients but Piron is well aware that advanced analytics cannot be sold as a turnkey solution.

Many enterprise companies do not have this ideal infrastructure and lack the resources to build it quickly. This reality is what makes the service and product approach attractive for small AI firms to pitch. By simplifying the integration, companies like Dessa or, fellow Toronto firm, Rubikloud can achieve a small amount of lock-in when onboarding an enterprise client.

And that lock-in — no matter how small — will make the race for market penetration interesting in the next 1–2 years. Rubikloud’s approach of focusing on a single industry (retail) is one way to get ahead. By ceding other opportunities (ie. the financial sector) to its competition, they can focus their solutions and their brand in one sector.

There will also be an interesting push and pull as enterprises themselves try to balance which projects to dole out to the AI firms and which ones to keep in their back pockets for their own growing data science teams. For enterprises more mature in their analytics journey, the movements and solutions offered by AI startups can provide a benchmark for up-to-date applications.

On visualization

“Command Centre” is not a term many companies get to use as a descriptor of a room in their workplace.

In Panel 1 of ABD 2018, Dan Zikovitz @ GE recounted the building of the Humber River Hospital’s “data wall” as a means of operationalizing the hospital’s investments in advanced analytics. It is an opaque but extremely comprehensive UI for a plethora of analytical engines from multi-agent simulation models and several algorithms for multi-layered capacity planning problems. Here’s how it works in their own words:

Is the dashboard to end all dashboards the ultimate end-game of analytics?Eric Bogart of Acosta describes his firms’ attitude to dashboard-ing in a single word: allergic.

“Data visualization is a way to inspect the decisions we’re making at scale and communicate our value. But broad scale visualization is something that will consume more than it produces.” — Eric Bogart @ ABD 2018

Naturally, Bogart occupies a different space than HRH (retail not healthcare) where analytics can play a much different role based on the specific use case. That being said, it raises some interesting ideas about the strategy of knowing when and how to utilize data visualization.

Data viz is a staple part of many data analysts’ day-to-day activity — whether by habit or mandate. If inspecting a chart you made would lead you to make the same decision you were going to make anyway, why make the chart?

This line of thinking is really a question of efficiency. It’s not that the chart has no value — it’s that the time spent making it was traded for time spent analyzing something else. Human capital is expensive, so what if that chart was automated?

Yes, it’s a small thing. But a single glance at a header of distributions can give a team of analysts a useful at-a-glance view of something they manipulate every day. Enterprise organizations are best poised to benefit from these tiny design efficiencies at scale. With good data science talent being so expensive, visualization as tooling is one small way to make sure the experts you’re hiring are doing what you’re paying them to do.

Consider what kind of interaction best suits the end user of a built and trained analytical process. An interactive dashboard? Or a single push notification to the key stakeholder? Ideal design of an analytics solution considers the process from the data generating mechanism all the way to the ultimate end user(s).

End-to-end design is a luxury

Enterprise analytics as a process, might look something like this:

1. Data Generating Process
2. Cleaning / Feature Engineering
3. Analytics Engine
4. User Interface
5. Decision

Being able to design and optimize every stage of this workflow is a luxury that enterprise organizations don’t always have. Additionally, every stage can look different for every company and every stage has has its own pitfalls. Here’s two that we’ve heard come up several times:

Attributing a failure to derive an impactful decision at the end as the sole failure of the analytics engine.

Have a clear measurement strategy for your approach and a strong preliminary understanding of your data.

Conduct a thorough data exploration — not only on the data itself, but also audit the process that generates it. Outline all the caveats about the process with respect to the inferences you’re hoping to make once the model is in production. This will help guide your measurement strategy and most importantly, help you identify the minimum amount of value you can get.

In other words, is even the “dumbest” version of your model, still providing some amount of incremental value?

Data cleanliness may be mitigated by data engineering efforts but data availability is a business and strategy problem.

Data enrichment as a strategic consideration when making business decisions is often not something large enterprises are actively doing (yet). In fact, many organizations struggle to find value in the data they do have.

• What are the other processes adjacent to or related to the one which you are sourcing your data? Can you access those?
• For product-driven companies: how can you design features / encourage new forms of interaction to get more robust info from your user base?

Parting thoughts on technical debt in ML processes

There is one last thing we hear often within the discussions of analytics for enterprise. That is the notion of the quick win project to help kickstart analytics. We’ll be diving deeper into this topic with future articles. But for now, we just want to highlight one concept: technical debt.

For those not familiar with the software engineering concept, it’s relevance to ML and AI is important. The idea is that agility in engineering often comes at the cost of technical debt. Speed now, begets future (more painful) work. There are strategies and frameworks in software engineering to mitigate technical debt but many of them don’t account for the host of ML/AI specific debt that comes from advanced analytics processes.

If you’re an analytics professional in a large enterprise trying to secure buy-in for future analytics investment, be aware of moving too quickly or taking your quick-win project farther than it needs to (or should go). It’s perfectly ok if your companies first foray into advanced analytics gets scrapped before seeing widespread deployment. Even if you have to go back to the drawing board, if you’re doing so with better and proper resources than when you started that first project, that’s really a win.

For curious readers, we highly recommend the paper “Hidden Technical Debt in Machine Learning Systems” which serves as an excellent primer into the details of what makes technical debt exponentially tricky with ML.

Analytics By Design is a Toronto based community of analytics practitioners looking to understand the design, development and implementation of ML/AI strategy and technology.

Follow us on Twitter to stay up to date on new articles! And check out our website for more content.

AI for Enterprise was originally published in Analytics By Design on Medium, where people are continuing the conversation by highlighting and responding to this story.

Learn about Nitesh’s career transition from academia to consulting

As a Deloitte Consulting alumna myself, the diversity of talent that enters the world of consulting never fails to surprise me. I can draw a parallel to the even more proliferated world of Data Science today. In fact, career transitions into Data Science and AI are becoming more and more common with the screeching demand for Data Scientists, Machine Learning Engineers, Data Analysts and the like.

Nitesh Soni, Lead Data Scientist at Omnia AI — Deloitte Canada’s AI practice, is a superb example of talent that manoeuvred a transition from academia. During our initial conversations at the Analytics by Design (ABD) Inaugural Conference, we learned that Nitesh was a physics researcher in his former life and made direct contributions to discovering the Higgs boson. In contrast to the client-facing nature of his role in consulting today, Nitesh’s career change may seem challenging to some. We made it a mission to learn about his journey and to inspire those that might be contemplating a similar move.

Our correspondents, Aaron Zhang and Michelle Liu, sat with Nitesh for an interview late November, during which he touched upon the challenges that new entrants from academia may face when transitioning into the corporate world. He also elaborated on the ways academic research and practical data science diverge and intersect.

ABD Correspondents: Nitesh, thank you for taking the time to chat! To start, tell me about yourself and how you ended up in Data Science and AI Consulting.

Nitesh: Thank you for inviting me! Before I talk about my current role, I can share a little bit of my personal story.

I grew up in a small town in northern India. I always loved science and mathematics. Just like a typical science geek, I always strived for academic excellence throughout my education — like breaking the state record by scoring 100% in my undergrad math class! Fast forward to my PhD studies, I worked on the Belle Experiment at Laboratory KEK in Japan with a group of world-renowned researchers including some Nobel laureates.

Thereafter, I held multiple post-doctoral research positions at universities in UK, Canada, Australia and earned the opportunity to work at top ranked research laboratories like CERN in Switzerland and SLAC in the US. All of my research work was heavily focused on experiment design, data collection and analysis, which involved manipulating massive data sets and extracting rare insights from millions of data points. Naturally, I became immersed in the world of Data Science.

ABD Correspondents: Which research project stood out the most?

Nitesh: I would say the research that led to the discovery of Higgs boson — one of the biggest scientific discoveries of this century in particle physics. I made direct contributions as a part of the ATLAS collaboration at CERN.

ABD Correspondents: With already so much success in academia, what made you want to venture into the corporate world?

Nitesh: That’s a great question. I was relatively comfortable in research and had a lot of success — received a few prestigious awards including one from the Department of Science and Technology in India and had the opportunity to be part of the Lindau Nobel Laureate Meetings in Germany.

But after the discovery of Higgs boson, I was thinking that such discoveries rarely happen in quick succession. Though it is exciting to contribute in ground-breaking research, a discovery of this magnitude might not happen again within the tenure of my career in academia. This is when I started thinking about a path outside of research. I thought: let me challenge myself and venture into the corporate world where I can solve real world problems. Fortunately for me, the move worked well.

ABD Correspondents: What are the major differences between the nature of work in research and in the corporate environment?

Nitesh: The corporate world has many stakeholders. They need to focus on quick wins and short-term goals. So when selecting and implementing analytics projects, businesses may prioritize projects that realize business value quickly at the cost of reduced complexity.

Research roles require more in-depth knowledge of the subject at hand. Effort is invested in building or researching something that either does not exist or exists with unproven accuracy. In most cases, the impact of research work is not realized immediately or even in the near future. As an example, electrons were discovered in the 19th century and, at the time, no one would have thought that it was going to have such a huge impact on civilization until the 20th and 21st century. Essentially, every electronics gadget evolves around the mechanics of electrons in some way.

That being said, with more businesses starving for analytics talent, entry-level data scientists have a better chance of getting a corporate job more often than an academic one.

Lastly, there was one thing that definitely took me some time to get used to — the dress code! The most formal attire for me as a researcher back then was a blazer and jeans, and I would only wear them when attending important academic conferences. Whereas in consulting, suit-and-tie is the standard.

ABD Correspondents: What are the main challenges data scientists from strong academic backgrounds face when they enter the corporate world?

Nitesh: In the world of research, you are always surrounded by like-minded individuals who communicate in a similar language. However, in the corporate world, we interact with people from very diverse backgrounds and various levels of technical aptitude. I find that researchers often have a hard time delivering a concise message without getting into technical details. It is important for a data scientist to adjust their communication style according to the audience in the corporate setting.

The second challenge comes from the misalignment between personal interests and organizational goals. In a company, the outcomes of projects need to align directly with strategic objectives that are determined by the organization’s top leadership. As a researcher, I felt that I had more control over the projects I wanted to work on. To adjust to this misalignment, a data scientist from academia should learn to grasp the business context from the get-go and to apply analytical insights in a way that can influence the direction of the project.

ABD Correspondents: As an emerging field, there are many entrants looking to break into the industry. What’s your advice for those just starting out?

Nitesh: First of all, broaden your technical skills and immerse yourself in the world of Data Science. There are tons of information accessible online where virtually anyone can start learning. Though, technical aptitude is definitely important.

Secondly, build your network and let your curiosity guide you towards where you want to be! Attending industry workshops and seminars is a great way to develop a basic understanding of the field.

ABD Correspondents: Any specific advice for entrants transitioning from academia?

Nitesh: I think it is important to clarify the role you are looking for and to understand the types of positions that are suitable for you. There are many roles that do not require deep technical skills. For instance, if you are interested in project management, you likely will not need an advanced degree in machine learning, a certification in the subject area might suffice.

I often meet people that enter the industry without prior understanding of what the role entails. It is frightening and frustrating, not only for these individuals but also for the employer. We met a PhD candidate who applied for a consulting role and expected to do only hands-on coding, but in reality, a consultant needs to do much more than that. It is important to know what you want and to be transparent about your requirements with the employer. I think this mindset will help bridge the gap between the skills you have and the expectations of the company you want to apply for.

ABD Correspondents: All great points! Thanks again for taking the time to chat.

Nitesh: Of course — thank you for having me.

Want to learn more about careers in Data Science? Follow Analytics By Design and stay tuned for similar conversations with working practitioners in Advanced Analytics, Data Science and AI.

Analytics By Design (ABD) Toronto

Omnia AI, Deloitte’s artificial intelligence practice leads all others in starting, enabling, accelerating, and sustaining the AI journey. Check out open roles at Omnia AI — Deloitte Canada’s AI practice!

Breaking into Data Science: From discovering the Higgs boson to AI Consulting — Nitesh Soni was originally published in Analytics By Design on Medium, where people are continuing the conversation by highlighting and responding to this story.

In May 2018, Analytics by Design (ABD) welcomed over 350 industry practitioners, business leaders and scholars at the Art Gallery of Ontario for its inaugural conference. The one-day event focused on the impacts emerging technologies like artificial intelligence (AI), machine learning (ML), big data, blockchain and internet of things (IoT) would have on businesses, government and society at large.

Designing AI for the Enterprise with Stephen Piron

Stephen Piron kicked-off the conference with a thoughtful discussion around the challenges of designing AI for enterprise. Piron demonstrated the common refrain of “analytics challenges being managerial” at implementation stages when discussing stories of use cases Dessa has tackled in the past. To set up the proper infrastructure for ML and AI, he said that AI firms will need to be effective social engineers, capable of building sponsors across many teams in large organizations to succeed.

The State of AI with Lovell Hodge

Lovell Hodge, Vice President at TD, brought the audience back to where AI started, outlining the key findings throughout the discovery and research journey of AI and discussed where we are heading as humans. Hodge believes that AI is a multi-disciplinary field with the goal of mimicking intelligent human properties such as reasoning, adaptation and learning. He discussed many popular conversational AI platforms (Siri and Google Home) which are becoming more mature and changing the human-machine interface. Missed out? Catch the full talk here:

Panel 1: The Analytics Development Life Cycle
Our first panel shared their different experiences of leading an organization’s analytics journey. All panelists mentioned the challenge of dirty data and had a lively chat on how and when to use visualization effectively in an analytics process.

Panel 2: Breaking Barriers & Unleashing the Power of Analytics
Panel 2 took a deep dive into analytics strategies and platforms, weighing in on how to structure analytics teams in big firms and small start-ups. As the industry matures, the panel spoke to the emerging trends around user interaction with data and analytics tooling and the speed at which the field was becoming more accessible. Yeloglu said that even the simplest applications of AI are optimizing the way we work and are in fact having a big impact on our society. She is excited about AI for Good, specifically empowering socially-disadvantaged individuals (i.e. applications for AI for the visually impaired). Missed out? Catch the full talk here:

Keynote 4: Our Digital Twins & the Trustless Economy with Gary Schwartz
Gary Schwartz opened the afternoon with some refreshing ideas around blockchain technology, illustrating its potential impact on our notions of identity. Schwartz captured the audience’s’ attention while discussing heady ideas around how society trades in trust. He suggests that as our digital footprints via social networks continue to expand, it becomes more and more challenging to maintain sovereignty over our identities and points to the latest developments at Facebook and more as key to the current zeitgeist. Schwartz identifies blockchain technology as being a pivotal innovation that is poised to redefine this whole conversation.

Keynote 4: Living in an Age of Disruption with Douglas Heintzman
Our 4th keynote speaker, Heintzman inspired the audience with the possibilities that the digital age has to offer but from a macro-level. As the final keynote, Heintzman provided a cool hype-free counter perspective to technology as something on the periphery of larger societal changes. He expanded the scope of the day and reminded the audience of the larger scale impacts outside of business. Heintzman challenged today’s leaders to think through the economic, political and societal implications of automation, artificial intelligence and globalization and spoke passionately about the potential of Toronto specifically to succeed in the face of all the emerging technologies covered throughout the day.

Panel 3: Governance, Regulation and Growth in the Digital Age
Our third panel spoke to the ethical implications and regulations of the digital age. With an enormous amount of data at our disposal, it is important to protect customer’s information — especially at the banks. Instead of viewing governance and regulations as the enemy of innovation, DosSantos expressed that the right level of governance would in fact allow developers to access data faster and more effectively.

Panel 4: The Convergence of Artificial Intelligence, Machine Learning, Robotic Process Automation and Internet of Things
Jenny Hui (Element AI), Simon Smith (BenchSci), Alex Mohelsky (EY) and Tom Emrich (Super Ventures) brought their expertise in AI, Healthcare, Technology and Augmented Reality to discuss the converging trends in these technologies, such as robotics, IoT and AR. This convergence is paving the way for more exciting, sustainable analytical solutions, but it is equally important to balance realistic expectations and creativity when building the next novel application. All panelists agreed that many companies have a fear of missing out when it comes to technology.

Panel 5: Blockchain: Hype vs. Reality
For our final panel we welcomed back our afternoon keynotes and a host of other blockchain practitioners to discuss the applications of the technology beyond crypto currencies. Much of the panel was focused on a single topic of discussion: “what the killer app of blockchain” would be — collectively the panelists engaged in a fiery discussion bouncing a host of ideas back and forth. They covered: empowering citizens through sovereign identity, increasing operational efficiencies in large institutions, bitcoin, digital right sharing and more. Despite the exhaustive list, they all believed the best has yet to come.

Fireside Chat — Building Up the Next Generation of Analytics Talent
We ended the day with Mark Wagner (Scotiabank) and Professor Stephen Thomas (Queen’s University) discussing the skills — both technical and managerial — that the next generation of analytics talent would need to succeed. Wagner highlighted organizational differences across industries and emphasized the importance of being bold and courageous in recruiting to retain top talent.

Workshops and Case Competition. In conjunction with a full-day of panels and keynotes, the ABD conference hosted two technical workshops in partnership with We Could Data and SAS, and a case competition where over 30 teams of 150 contestants were given a real transactional dataset by Rubikloud to solve a real business problem in the retail domain.

It was really difficult to cram a full day’s worth of discussion into a single article without short changing all of the incredible discussion! The whole ABD team is looking forward to 2019 and diving deeper into the stories, strategies and technologies of the future and we hope you’ll join us as we do so :)

Stay up to date with us on our website: https://abdtoronto.com/

Thanks to our strategic partner, Smith School of Business (Queen’s University) as well as our event sponsors TD, Scotiabank, LoyaltyOne, Dunin-Deshpande Queens Innovation Centre, Palomino, SAS, and our community partners, Rubikloud, We Cloud Data, ICF Olson, ThinkWire and Smith Women in Analytics.

Analytics By Design 2018 was originally published in Analytics By Design on Medium, where people are continuing the conversation by highlighting and responding to this story.