Understanding How EV Charging Behavior Affects Distribution Networks

The International Energy Agency projects that 30% of all vehicles will be electric vehicles (EV) by 2030. This transition, at the intersection of electric power and mobility, combined with increased generation from renewable resources has the potential to significantly reduce greenhouse gas emissions in the years ahead. To make this happen, utilities who operate the distribution network need to understand how this new demand for electricity will affect smart grid assets. Our primary job at VIA is to help utilities navigate these shifts by understanding their data and fostering collaboration through our Global Data Asset Collaborative™ (GDAC™) program. As an example, VIA recently kicked off our first GDAC™ by focusing on transformers. Through this GDAC™, we are beginning to see that transformers are stressed by the switch to EVs and our focus will be on helping utilities find ways to keep these assets healthy over the coming years.

There are at least two things that make charging an EV different than, say, running a central AC unit. First, the power that needs to be delivered to an EV is around 20kW, which is four or five times the power required for a typical central AC unit, which ranges from 3-5kW. A “short-range” charge to power the EV so that its owner can commute could require around 40kWh, thus a “slow-charge” for a “short-range” car requires about two hours of charging. Powering large fleets of EVs will clearly require extending the capacity of current electricity distribution networks. 

The second issue that makes charging an EV different is timing. The timing of EV charging events changes the daily load profile of the home, workplace, and in urban centers equipped with networks of charging stations. Transformers are generally able to run past their rated capacity so long as they are given ample time to cool overnight. That is changing as commuters return home after work to charge their vehicles, never allowing transformers that time to cool down, which can cause them to malfunction and in extreme cases, explode. EV charging events, because they demand so much power so quickly from the grid, can lead to shifts in voltages along the distribution network. This leads to wear and tear on tap changers and other voltage regulation mechanisms. 

Utility asset managers need to understand which transformers in their fleet are most at risk as EV penetration increases. A recent study by researchers at Ohio State University illustrates what needs to be done to understand the effects EVs have on transformers and voltage regulators (“An Integrated Algorithm for Evaluating Plug-in Electric Vehicle’s Impact of the State of Power Grid Assets”). The authors have studied a representative sample of urban, suburban, and rural areas and tried to answer the question “What would happen to the distribution grid if each home had an EV?” To understand both the total load and the rapid charging behavior, the authors used actual distribution grid topology provided by American Electric Power (AEP) and simulated the behavior of the system as EV charging events are inserted into today’s “baseline” load demand. The authors find that suburban areas are expected to see the greatest stress, as it is assumed that, in urban areas, additional power will be provisioned by specific “fast-charging” stations while the suburban dwellers load will stress the transformers that serve their primary residences. In rural areas, the lower population density typically means that the transformers are not as heavily loaded as in a suburban area. Some authors predict long-term changes in mobility patterns that will increase the number of rideshare services (i.e., Uber). Rideshare cars are typically required to drive all day and would require longer charge times. This corresponds to the most aggressive scenario studied by the authors, in which case they expect insulator degradation to occur after just one year. The results illustrate the socio-technical complexities of planning the future smart grid and the need for detailed studies on how people are expected to use their vehicles.

Author’s Note

As a highly-trained problem solver with deep scientific and computing expertise, I’m always hungry for tough problems to solve. There’s no doubt that integrating EVs into the smart grid is a tough problem. More importantly, it is a high-impact socio-technical problem that we as a society need to solve to transition to a greener future. Working together with the world’s largest utilities, VIA is in a position to help solve these problems, a privilege I am grateful for every day I go to work. At VIA, we have a company value, “Love in=Love out” which means that if you love what you are doing, you will do great work. I expect we will do great work in this area, and help our customers navigate the challenges of the EV revolution.

The Importance of Unit Testing

In March, VIA’s Ashley DaSilva, Team Leader, Product Development, was invited to lead a workshop on unit testing for McGill’s Computer Science Graduate Society. The workshop was part of their seminar series: CS Tools and Tricks, which introduces graduate students to topics they may not otherwise explore in depth in their academic programs. Ashley discussed the critical importance of software testing, why developers should embrace unit testing, and when and how to use mocking. She shares her experience with unit testing and a recap of her workshop below.

Learning Never Goes Out of Style

I first learned the importance of software testing in the early days of my theoretical physics Phd program. Back then, most of my coding was limited to scripting. I wrote scripts to model physical systems, analyze data, and visualize results. Over time, I started writing modules to be reused by myself or my colleagues for different projects. The first time I attempted to refactor one of these modules, it broke in unexpected ways, and I spent days tracking down and resolving all the issues.

I’ve grown significantly as a developer since those days, and now lead a product development team at VIA, focused specifically on the Trusted Analytics Chain™ (TAC™). Every day, my team and I build, test, and deploy Docker containers with microservices. This includes Airflow and RabbitMQ for scheduling tasks, Redis as a cache storage, and BigchainDB to host a blockchain. Software testing is critical to each stage of product development, and something we constantly work to improve.

Unit testing is the foundation of VIA’s software testing process, and an essential skill for all of our developers. For example, TAC™ contains several components that all need to communicate with each other. We maintain a list of python scripts that are authorized to run on the system. Components of TAC™ must download this list to verify the checksum of the scripts. If we did not use mocking for the content of the list in the unit tests of these verification functions, then every time the list was updated, the tests would all have to be updated to account for the change. With mocking, we are free to update the list of scripts without affecting the status of the unit tests.

Ready, Set, Resilience!

Clean Architectures in Python by Leonardo Giordani is a great resource for learning more about unit testing and test-driven development. During my workshop at McGill, I presented examples of unit testing and mocking and shared a few exercises for the students to practice on their own. Some of these exercises came from the github repository associated with Giordani’s book. I’ve included some other examples below:

First, let’s look at a snippet of code. The code below shows a DataAnalyzer class. It has a method, get_data, which is a placeholder for however one would want to retrieve the data from an external resource (e.g., a database or an http request). It also has a method,
analyze_data which performs the sum of the items in a list:

class DataAnalyzer:
   def get_data(self):
      # Gets data from an external resource
      pass
      def analyze_data(self):
         data = self.get_data()
         result = sum(data)
         return result

The code below shows one example of a unit test that uses mocking of the get_data method:

from unittest import mock
from calc.analyzer import DataAnalyzer
def test_analyzer():
   analyzer = DataAnalyzer()
   with mock.patch("calc.analyzer.DataAnalyzer.get_data", return_value=[1.0, 2.1, 3.5]):
      result = analyzer.analyze_data()
   assert result == 6.6

In this example, the unittest.mock.patch will be applied to the method specified as its first argument and return the assigned return value every time that method is called from inside the scope of the patch. A sample list is assigned to the return value of the patch of the get_data method. This list should have the same format as the expected output of the get_data method, which in this case is a list of floating point numbers. Finally, the result of the analyze_data method is checked that it matches the expected value.

Mocking can also be used to check how you handle exceptions:

import pytest
from unittest import mock
from calc.analyzer import DataAnalyzer
def test_analyzer_connection_error():
   analyzer = DataAnalyzer()
   with mock.patch("calc.analyzer.DataAnalyzer.get_data", side_effect=ConnectionError("Could not connect.")):
      with pytest.raises(ConnectionError):
         analyzer.analyze_data()

In this example, instead of specifying a return value, there is a side effect. When the specified method is called, the side effect will be executed. In this case, a ConnectionError is raised by the get_data method. Using a side effect is particularly applicable when you have logic in your code that catches and recovers from errors.

Stay Curious

I enjoyed leading a thoughtful discussion on unit testing, and fielded some great questions from the students. One that stood out to me was:

“How can developers make sure that their mocks don’t get out of date?”

This is a really important and sometimes tricky topic! At VIA, we know that unit testing is only the first step of software testing. We also use other types of tests, like integration tests or end-to-end tests help identify problems in mocking before our software reaches users. And isolating and resolving problems at that stage is key to setting ourselves up for a successful integration. 

Mocking allows the freedom to isolate one particular part of your code and focus your unit tests on that functionality. Ideally, the expected inputs and outputs of the component being mocked are not going to change. This is typically true for external modules that you will use, at least within a particular major release of the software. However, if you know a software update will cause a change to your internal code base, it is your responsibility to recognize and communicate how that will affect your teammates. That’s why VIA believes that, in addition to testing, strong team communication and values like Learning Never Goes out of Style, Ready, Set, Resilience!, and Stay Curious are what help us develop and deliver the best iterations of our software to our users. 

The 5 Game Changers that Made 2018 VIA’s Year

Trusted Analytics Chain™ (TAC™)
In 2018, VIA’s Trusted Analytics Chain™ (TAC™) moved from early development stages to first pilots and now, is ready for an official launch in Q1 2019. Along the way, we were invited to speak about TAC™ at events across three continents, including the 9th Asian Leadership Conference in Seoul, Iberdrola’s Innoday in Madrid, and Greentech Media’s Blockchain in Energy Forum in San Francisco.

We were honored to accept the MITX Best Technology Innovation Concept award, which recognized TAC™’s potential to transform the energy industry. In addition, we demonstrated TAC™’s capabilities during our first ever webinar this past fall.

CEO Colin Gounden ALC

VIA TAC MITX Award

VIA Greentech Media

 

 

 

 

 

 

Team Growth and Development
VIA nearly doubled in size this year with the addition of 10 new team members in both our Somerville and Montreal offices. And, we are actively recruiting for the following positions to be based out of the Montreal office: Front-End Developer, User Interface and Experience Designer, Software Developer, Software Engineer, and DevOps Specialist. Visit our Careers page for more information and to apply.

VIA Hiring

Press Features
VIA enjoyed a record number of press features this year, highlighting both our technology and the team behind the tech. In January, CEO Colin Gounden was interviewed for Inc. Magazine to discuss VIA’s approach to building its team.

Additional media outlets include: com! Professional Magazine, FutureTech podcast, Digitex Futures, and Clean Energy Finance Forum. For links to these features, visit our Press page.

VIA Inc. Magazine

 

 

Brand Refresh
In March, VIA debuted its new name, logo, and website, as part of a brand refresh. We also introduced “Solve with VIA” as an anchor to our brand, one developed through brainstorming with the entire VIA team and consulting with our most trusted partners. The idea that clients “Solve with VIA” was an ever-present theme through these creative sessions, and ultimately inspired VIA’s newly designed logo. The logo visually represents the journey VIA takes with its clients from identifying a problem to finding a solution.

 

 

 

 

 

 

Partnerships
We are so grateful to our partners that have helped support our progress and made our success possible. And, today we are proud to announce our newest partner, KWHCoin. Through our partnership, KWHCoin will use VIA’s Trusted Analytics Chain™ (TAC™) to securely and anonymously analyze consumer behavior from smart meter data in order to help its utility partners better incentivize their customers to use renewable sources and clean energy. Earlier this year, we also began working with BigchainDB, leveraging their database to make TAC™ a consortium blockchain.

We were accepted to Accelerace, one of Europe’s top-seed accelerators, as part of their Cleantech program. Accelerace will help us establish a corporate presence in Denmark, make introductions to leading European utilities, and gain mentorship from experienced Danish entrepreneurs, cleantech executives, and industry experts.

Additionally, we were accepted to NVIDIA’s Inception program, which is designed to nurture startups revolutionizing industries with advancements in AI and data science.

VIA Accelerace

VIA Spotlight: Girard Newkirk, Founder and CEO of KWHCoin

Girard Newkirk, Founder and CEO of KWHCoin, spoke with VIA about his journey from retail executive to founder of a decentralized virtual global power company. KWHCoin aims to deliver reliable, clean, renewable energy access for disadvantaged and underserved communities across the globe using blockchain technology and smart contracts.

Tell us about yourself.

I was born and raised in Pender County, North Carolina and attended East Carolina University, where I studied Political Science and Finance. I have two beautiful children: a six-year-old son and four-year-old daughter. My professional background is in retail, and most recently, I worked as a Vice President for Macy’s in the Silicon Valley District.

Prior to founding KWHCoin, you were a retail executive for over a decade. What inspired you to leave that industry and pursue building this platform for renewable energy resources?

I have always been interested in finance, energy, and building systems to provide access and social impact for the disadvantaged and underserved. But the real turning point for me was my daughter’s stroke. She had a stroke at birth and struggled to survive. This experience put into perspective the need to live a purposeful life. As she lay in the NICU, I promised her that I would work on something to make the world a better place for her and my son and so, here we are.

KWHCoin aims to “improve the lives of the 1.2 billion people across the globe without reliable energy access.” What are some of the key ways KWHCoin works to accomplish that mission?

As we began to travel to and interact with markets in Africa and the Caribbean, we quickly learned that the solution needed was more than just energy trading and a blockchain. We discovered we had to develop an entire ecosystem to support the needs and overcome current barriers that kept 500 million households without energy. So, we modified our original plan. Now we have become an energy solutions provider as well as a platform for sustainable infrastructure development.

You talk about the “Internet of Energy,” and how the internet can act as a digital rail to deliver clean, localized energy to members of the network. Tell us about your vision for the global impact of an Internet of Energy.

I think the energy generation and distribution systems of the very near future will be much smaller, localized, and driven by both demand response and the coordination of software for efficient delivery. We envision all distributed energy resources as being critical to grid reliance, so we decided to design our platform as a mechanism for distributed energy resources to collaborate and communicate in a transparent environment. This enables each energy resource to become its own energy company and this independence will flourish in the coming years.

VIA uses its blockchain-based Trusted Analytics Chain (TAC) to establish secure access to confidential and distributed energy data, making it available for AI initiatives like predictive maintenance. How do you see TAC, or similar applications, fitting into KWHCoin’s goal of establishing a decentralized virtual global power company?

I see TAC as being a critical component to the development of the Internet of Energy. Security and coordination are the hardest and most important elements to the success of decentralized energy development and TAC is a bridge to connect data, energy usage, mitigate security threats, and foster collaboration.

The United Nations’ scientific panel released a report in October describing a strong risk of climate crisis as early as 2040. How do you think KWHCoin’s work is helping to combat some of the contributing factors of climate change? What further action do you think needs to be taken to prevent some of the more severe consequences of climate change?

I think we immediately need to invest in sustainable infrastructure development and work together to build compensation models that will attract investment and reward the public for good behaviors that contribute to the deployment of more renewable energy sources.

We often describe blockchain now as the internet in the early 90’s: it’s difficult to imagine the impact it will have on everyday life. That being said, how do you think blockchain will influence the way we live in the next 5 – 10 years?

I think we will see an array of disintermediation. The public, if we execute distributed systems and technologies effectively, will benefit tremendously from lower cost and more security of their data.

Would you like to add anything else?

Thank you for this opportunity. We encourage everyone to check out KWHCoin and join us on our mission to electrify billions and build a more sustainable planet.

So, You’ve Updated Your Privacy Policy

If you’re like me, your inbox has been flooded since late April with emails announcing updates to privacy policies and terms of service. Consumer brands like Yelp and Etsy, social media platforms like Twitter and Instagram, and, it seems, everyone in between all sent emails. It’s no coincidence. The newly implemented General Data Protection Regulation (GDPR) means any company operating in the European Union (EU) needs to update its existing privacy policies to comply with this legislation. And as it turns out, that’s quite a lot of companies.

While it was specifically passed in the EU, GDPR has far-reaching implications for companies around the world. The policy addresses consumers’ growing need for transparency and privacy, which, at first glance, seem to be competing interests. But WIRED offers this helpful explanation: GDPR “gives people the right to ask companies how their personal data is collected”, stored, and being used, and to “request that personal data be deleted. It also requires that companies […] get your consent before collecting it.”

Consumers’ growing need for greater data privacy, security, and anonymity, as well as transparency from companies that collect data, extends to the energy industry. Smart meters are a great example of this: EU Member States “have committed to rolling out close to 200 million smart meters for electricity and 45 million for gas by 2020″, at which time it is “expected that almost 72% of European consumers will have a smart meter for electricity while 40% will have one for gas.”

In theory, utilities could leverage the massive volume of consumer data collected by these meters for AI initiatives that seek to improve predictive maintenance and energy efficiency. However, utilities don’t own this data, consumers do. And the EU wants to safeguard it. So, how can utilities balance consumers’ rights to data privacy, security, and anonymity, while leveraging this wealth of new information to improve service reliability and efficiency?

At VIA, we are enhancing our Trusted Analytics Chain (TAC) platform to address this challenge. TAC Permissions is a new feature that allows energy companies’ customers to determine who has access to their personal data and how it is used. Permissions are stored in TAC’s blockchain, making them immutable and auditable by consumers, corporate compliance departments, and government agencies. VIA’s solution helps to alleviate the tension between:

  • Customers’ needs to protect their data
  • Energy companies’ and algorithm providers’ needs to access data to develop and implement valuable services such as improved energy efficiency and reliability.

Smart meters are just one example of the many digitalization technologies that energy companies are using to transform their operations. The industry as a whole is more data-dependent than ever before. In parallel, consumers are more aware than ever of the consequences associated with misuse of their personal data. Legislation like GDPR requires companies to know and comply with customers’ preferences about using personal data. VIA is proud to offer TAC Permissions as a solution for ensuring that customers’ wishes are respected while AI initiatives yield new ways to improve service and efficiency.

Six Values in Six Months: A Co-op Reflects

My name is Meg Foley. I have worked as the marketing execution co-op at VIA’s Davis Square headquarters since January. And this is the story of how that became the most rewarding experience of my life (so far!).

That’s me!

In my two years at Northeastern University, I have already made major progress towards some of my life goals, like traveling the world, earning a college degree, and getting real-life work experience (and I still have three more years to go!). I studied abroad in Dublin my first semester, went to classes in Boston for the next two semesters, and worked at a startup in Davis Square for the past six months.

These experiences have all made me who I am today and my time at VIA is no exception. I gained essential professional experience, and each lesson stemmed from VIA’s company values.

Respect a good challenge and challenge with respect

We believe in facing challenges head-on with passion and excitement. We’ve created an environment where team members feel psychologically safe to raise alternative ideas and share their personal concerns, helping us reach better outcomes in the end.

The team loves to step out for ice cream on a hot day! And after a particularly warm day in February, J.P. Licks even featured one of our ice cream photos on their social media.

As the marketing execution co-op, one of my primary responsibilities was maintaining our Zoho CRM (a database that helps us manage all our relationships with our ever-expanding network). But more than just maintain this system, I made it my mission to make it even more efficient and effective. One component of that meant making data reports, like our sales pipeline, more visually appealing. To do this, I worked closely with VIA COO, Kate Ravanis, to hash out the goals and strategy for how this would work.

For me, the biggest challenge was putting my Excel knowledge to work and being patient with changing formats and preferences. Through many versions and rounds of feedback with VIA’s executive team, I eventually found a way to incorporate everyone’s input and all the essential details. Now that visual report is a part of the team’s biweekly demo, and this challenging project led to a new staple piece of internal communications!

Learning never goes out of style

Learning is all about feedback and trying new things. Each person, at every position in the company, has frequent feedback sessions with their manager to set goals and priorities, reflect on accomplishments, and discuss areas for improvement and additional support.

Somerville team at our brand launch (this was a huge deal to be a part of)

One of my friends, who is also on a co-op at another organization, was shocked to hear how often I speak with my supervisor on the marketing team. “I wish I could get any feedback at all!” he said. I was proud to have such an open line of communication, which has been so important and useful for my personal growth.

This open communication also meant I was able to speak up about projects I wanted to pursue, like becoming the VIA Culture Coordinator, and skills I wanted to develop, like written communication, so I was really able to shape my role at VIA to better fit me. With each new responsibility I took on, I created a comprehensive process guide so future co-ops can continue building on projects I started, like our new Instagram account. I hope these guides allow for many future smooth transitions from one co-op to another, and that it’s one way I can continue contributing to VIA’s learning-focused environment.

Be each other’s biggest fan

Everyone deserves a high five once in a while! The more we support each other, the more we encourage wide ranging contributions from diverse backgrounds to solve problems.

The marketing team treated me to a fun brunch on my birthday.

Every person on the team is recognized and celebrated for any number of reasons: professional accomplishments, personal victories, or even birthdays.

In fact, on my birthday I was taken to a lovely brunch by my marketing team. This is a fond memory I’ll treasure forever. Celebrating the individual, like we did on my birthday, is central to VIA’s culture. As the Culture Coordinator, I was proud to be a part of creating even more structure around what we celebrate and how, like birthdays (of course), work anniversaries, and fun holidays like Pi Day and Star Wars Day.

Ready, set, RESILIENCE!

We are solving problems that others have considered unsolvable. Difficult problems require patience and grit. Individuals require resilience to approach a major challenge and overcome all the smaller challenges along the way.

During my co-op, VIA achieved a lot of exciting milestones. One that stands out in particular was winning the MITX Best Technology Innovation Concept Award. We were thrilled to learn we were finalists, but in order to qualify for the next round of review, someone from VIA would need to present a demo of our blockchain-based Trusted Analytics Chain (TAC).

Due to the very busy travel schedules of our client and business development teams, the only person available on the night of the demo was a member of our marketing team. She needed to learn (in just one weekend) the inner workings of TAC and how to demo it, and prepare for any number of questions that might come up with the judges (I was happy to help her prepare by asking the most challenging questions I could think of, and many of those were actually asked on the night of!). Without her determination and ability to deal with high stakes situations, traits that come naturally to everyone on the VIA team, this feat couldn’t have been accomplished.

Love in = Love out

We believe that if you love what you do, it will show in the quality and productivity of your work. Twice a year we get together for a company-wide offsite where team members’ contributions are recognized and new ideas are discussed.

The team has grown from then to now, we have 20 employees as of June 2018.

We had an amazing company offsite this April. This experience was one of the times I felt most immersed in VIA’s love in=love out culture. For example, during the offsite we had an ongoing activity: a fun fact scavenger hunt. Everyone received a list of fun facts with blank spaces next to them, and our goal was to match the anonymous fun fact to the right person. You would not believe how much we committed to this scavenger hunt! I had so much fun getting to know everyone on the team through this activity, while enjoying some friendly competition (people even formed alliances to pool their answers!). It just goes to show that when people really love what they do, it shows, even in small moments of downtime during an all company meeting.

Stay curious

Technology today is moving at a super fast pace. Curious people who constantly explore new approaches and think outside the box help find the best possible solutions.

Some of the marketing team at a recruiting event at Northeastern.

AI and blockchain technology are two of the hottest buzzwords in the tech world (and beyond) right now. But, explaining what they mean exactly is not so simple. The marketing team stays curious about how we can better make writing about what we do more accessible, and our CEO Colin Gounden does a great job of sharing creative anecdotes and analogies to help us.

One example of that is his Van Halen story: the band’s 1982 world tour contract explicitly stated there shouldn’t be any brown M&Ms backstage. Colin explains how this seemingly bizarre request actually illustrates how Ethereum’s smart contracting functionality works. People’s eyes light up every time he explains this (a real A-ha! moment). Marketing was even able to use this to write one of our latest blogs: “Rock Science: How Van Halen Invented Smart Contracts”.

Goodbye and thank you!

That’s my team (on St.Patrick’s Day) rocking green.

I will miss VIA. It was an adventure, one I couldn’t have trekked without the team here in Somerville (shout out to Marketing!). It must seem cliché at this point, but there is not a single thing I would change about my co-op. Being at a company that cares about the individual as much as the company is rare, and I think the strong set of values is to thank for that.

Rock Science: How Van Halen Invented Smart Contracts

Since their debut in the late 1970s, Van Halen has become one of the best-selling bands of all time, selling over 56 million albums in the US alone. Their music defined a genre (and a generation), and inspired countless musicians to follow in their footsteps. But, their impact reaches much further than the music world. In fact, you could argue Van Halen’s reach extends even into the blockchain boom we’re seeing today. How so? Well, Van Halen sort of invented the smart contract. Let me explain.

By Colin Gounden

You may be familiar with one particular bit of lore from Van Halen’s long and storied career: the band’s brown M&M clause. In their 1982 world tour contract rider, which clocked in at an impressive 53 pages, Van Halen specified that in the bowl of M&Ms in their dressing room, there should be “ABSOLUTELY NO BROWN ONES.” Now, the rider also demands herring in sour cream and four cases of Schlitz Malt Liquor beer, so it’s easy to chalk up all this to diva-like antics. However, in his autobiography Crazy from the Heat, former frontman David Lee Roth shares insight into the genius behind this oddly specific request:

“Van Halen was the first band to take huge productions into tertiary, third-level markets. We’d pull up with nine eighteen-wheeler trucks, full of gear, where the standard was three trucks, max. And there were many, many technical errors — whether it was the girders couldn’t support the weight, or the flooring would sink in, or the doors weren’t big enough to move the gear through.”

Roth goes on to explain that because their equipment needed to be handled with such specific care in order to ensure the safety of band and audience members alike:

“[…] as a little test, in the technical aspect of the rider, it would say “Article 148: There will be fifteen amperage voltage sockets at twenty-foot spaces, evenly, providing nineteen amperes . . .” This kind of thing. And article number 126, in the middle of nowhere, was: “There will be no brown M&M’s in the backstage area, upon pain of forfeiture of the show, with full compensation.” So, when I would walk backstage, if I saw a brown M&M in that bowl . . . well, line-check the entire production. Guaranteed you’re going to arrive at a technical error. They didn’t read the contract. Guaranteed you’d run into a problem. Sometimes it would threaten to just destroy the whole show. Something like, literally, life-threatening.”

Not divas at all, Van Halen used this M&M clause as a way to test that their equipment, their band members, and their audience were safe. And their if/then logic (if there are brown M&Ms in the bowl, then the contract wasn’t read carefully) is similar to how the smart contracting functionality of blockchain platform Ethereum works as well.

Similar to traditional contracts, Ethereum smart contracts set specific rules around how users can interact with each other and exchange items of value (e.g., money, information, or property, like band equipment). What makes smart contracts different than their traditional counterparts is their ability to code these rules into a secure blockchain and automatically enforce them. So, transactions on Ethereum can only happen if the specific parameters within the smart contract are met. For example, if Van Halen’s contract rider existed on Ethereum, the band could rest assured knowing that their equipment had only been handled by those who had followed the exact parameters of their contract.

Let’s face it: we’re not all rockstars like Van Halen and blockchain isn’t exactly a consumer technology (at least, not yet), so who does smart contracting benefit? Well, for starters: high-stakes industries like energy that historically haven’t been able to implement AI initiatives because of data security concerns. AI needs lots of data to make predictions, so for companies with highly confidential data, like utilities, lack of secure access to data has been a major blocker for AI. Smart contracting has the potential to spark a massive AI revolution in the industry, and that’s a really big deal.

Let’s say Utility A wants to predict when their transformers might fail. To do this, they would need to provide AI Solution Provider B with data like the exact latitude and longitude of every transformer they operate. This kind of information is incredibly sensitive (and oftentimes, a matter of national security), so without a secure way to share it, utilities could not provide the data necessary for a solution provider to build a customized AI algorithm. However, Ethereum could provide a platform for blockchain-based applications that set specific restrictions around how data can be accessed. In that case, Utility A could set the rule: AI Solution Provider B can access Data C for Purpose D, thus protecting their data from malicious actors or requests.

The potential impact of AI in energy is extraordinary. From predictive maintenance to resiliency planning to grid modernization, AI can improve the reliability and efficiency of operations and service of utilities across the world. And while adopting new technology like blockchain can be daunting for any industry, think of Van Halen’s question from one of their most recent hits: When was the last time you did something for the first time?

VIA Reflects on its Largest Company Offsite Ever

In April, VIA brought its Somerville and Montreal teams together for our first company offsite of 2018 and our largest one ever (with 16 team members!). Offsites are essential to our company culture: they offer an uninterrupted time of sharing team wins, solving project challenges, and planning for the next big thing.

By VIA Careers

Once the Somerville team arrived at our Montreal office, we launched into our “unconference” – a relatively new format for our offsites. We found that moving away from formal presentations and towards open, free-form discussions made the overall offsite experience even more productive and engaging for everyone.

It’s not entirely unstructured, however! Team members submit topics beforehand about what they want to discuss, like best practices for communication, how to identify the right professional development opportunities, and the future of blockchain in energy and beyond. We vote as a group to finalize the list of topics at the start of the offsite.

VIA company offsite      VIA company offsite

CEO Colin Gounden and COO Kate Ravanis kicked off the unconference with leading the vote on discussion topics and handing out gifts. This time, each team member got a copy of the original Winnie the Pooh, by A.A. Milne, with a personalized note from Colin. You may remember from our last offsite blogWinnie the Pooh is special source of inspiration for us.

VIA company offsite      VIA company offsite

The open discussion format inspires even more collaboration across the technical and commercial teams, inviting fresh perspectives and big picture thinking.

VIA company offsite      VIA company offsite 

After our discussion sessions, we hit the links for a friendly round of (glow in the dark!) mini golf: an impressive display of athleticism from all. No offsite would be complete without this time to step away from the day to day and connect with teammates outside of work.VIA company offsite

In the coming months, we’ll continue to grow our Somerville and Montreal teams. If you’re interested in joining the fun, check out our careers page for the latest opportunities.

3 Ways VIA Prioritizes the Team Behind the Tech

VIA is not your typical tech startup. From our unique application of AI and blockchain, to our belief that humans + AI solve problems better together than either alone, VIA has certainly carved out its own niche in a crowded field of solutions providers. But according to our team, what really sets us apart is our people-first mentality. In order for our technology and solutions to stay cutting edge, we need to continue building, cultivating, and supporting the team that got us where we are today.

By Kate Ravanis

1. Building a great team starts with recruiting great team players.

Our CEO Colin Gounden said it best in his recent interview with Inc. Magazine: “What we want are people who are smart and motivated, and who are good at solving problems. You can teach programming and data science, but not innate things like motivation and problem-solving and creativity.”

What we are really looking for are individuals willing to learn, evolve with our technology, and roll up their sleeves to do whatever comes their way.

2. Cultivating a great team requires an effective feedback process that facilitates a regular dialogue around individual’s desires and abilities.

We know that each person on our team has a unique set of strengths, areas of expertise, and career goals. And, we know these things are subject to change over time. At VIA, we rely on our feedback process to ensure we stay in tune with each person’s desires and abilities, which is the first step to ensuring they continue to feel motivated and productive.

One key component of our feedback process is frequency: we schedule bi-monthly reviews (instead of the more traditional annual review cycle) to provide consistent opportunities for self-reflection, goal evaluation, and constructive feedback. To borrow from Winnie the Pooh: frequent reviews are a chance to “stop bumping for a moment” and reflect on what’s working, what isn’t, and what we can change (and how) to become more effective.

3. Supporting a great team means balancing your team’s desires and abilities with the company’s needs. Our allocations team has a goal of matching the Desire, Need, and Ability (DNA) of people and the organization over time.
Understanding the DNA of our team is only one piece of the puzzle. The real value (for both the individuals and the company) comes from putting that information into action. At VIA, our allocations team uses this knowledge to make thoughtful decisions around project and talent management. We strive to balance what an individual wants with the tasks the company needs completed. This means making an effort to create opportunities for team members to learn new skills and tackle new challenges (and not just assign tasks within their current wheelhouse), which is essential to long-term satisfaction and team morale.

The AI That Cried Wolf: How VIA Refines Algorithms

As energy companies start to explore AI solutions, we hear a recurring set of questions: How do your algorithms work? How much data do you need to make predictions? How do you measure the accuracy of your algorithms? With that in mind, we wanted to take an opportunity here to shed a little light on one of our most frequently asked questions.

By Colin Gounden

Does VIA use subject matter expertise to build models or does it rely solely on AI algorithms?

The short answer is both. We start by building AI algorithms with a combination of our client’s equipment data (age, location, equipment type) and add contextual information like pollution or weather data. We use this data to create initial predictions. These initial predictions are refined with input from our client’s internal subject matter experts. That’s right: our goal is to create software that works alongside human experts rather than replaces them.

How does our collaborative approach to refining the algorithm work? To explain, let’s take the example of an AI model trained to distinguish photos of huskies from photos of wolves. Initial algorithms had a hard time with this task. VIA’s key differentiator is a mathematical approach that extracts from the AI an “explanation” for each prediction. In this example, early results explained that the model would classify the subjects as wolves when there was snow in the photo. An obvious mistake for people but not obvious for a computer. Once the snow feature was removed, the algorithm’s accuracy improved more than three-fold. Similarly, we present an algorithm’s initial predictions and corresponding explanations to a client’s team of experts and de-prioritize selected features.

We see two big advantages from this approach. First, we remove any spurious correlations that an AI algorithm may be picking up. Second, we also gain buy-in from experts and users regarding the software’s predictions. Increasingly, experts have “algorithm aversion” where they don’t blindly trust black box predictions. The ability to have explanations and input into the algorithms builds credibility in the software and recommendations.

Via Science’s Top 8 Highlights of 2017

2017 was a busy year for Via Science, filled with major milestones and great team wins. We are proud to share some of our favorite moments with you, and look forward to another exciting year ahead!

By Via Science Marketing

8. PJM selected Via Science for a proof-of-concept.

We are honored to help PJM ensure their more than 65 million customers have safe and reliable service this winter and beyond.

7. Via Science joined Elemental Excelerator’s 2018 Demonstration Track.

We will receive up to $1M in funding to further develop our new blockchain technology for the energy industry.

6. TEPCO and Via Science launched a joint initiative.

The initiative will utilize analytics and machine learning in an effort to improve reliability, efficiency, and service for TEPCO’s 29 million customers.

5. Via Science won Startup Canada’s Innovation Award (Quebec).

The award recognized our leadership in using AI to solve large-scale high-stakes challenges across critical infrastructure.

4. We joined a panel at NECPUC to discuss AI in regulated industries.

CEO Colin Gounden addressed the importance of building trust and credibility with AI, and how we are working to do just that with our blockchain platform, due out in early 2018.

3. In July, we moved into our new headquarters in Somerville, MA.

And this fall, we hosted our first company offsite in the new space.

2. We travelled the globe to meet with energy industry leaders and influencers.

Our thought leadership in AI and energy has been featured in global publications across the industry.

1. We began refreshing the VIA brand and will continue in early 2018.

Our brand refresh reflects our focus on providing transformational AI solutions for the energy sector. Changes will include a new website and logo so, stay tuned!

Using AI to Rebuild Critical Infrastructure After Hurricanes

Millions are without power across Texas, Florida, and Puerto Rico as the second most costly hurricane season on record intensifies. Experts warn that extreme weather patterns like these are the new normal, raising concerns that their impact on critical infrastructure (and the communities they serve) will only worsen.

By Via Science Marketing

Critical infrastructure is particularly vulnerable to extreme weather, and widespread destruction following catastrophic storms makes rebuilding even more challenging. Despite even heroic efforts, utilities struggle to restore power and repair damaged infrastructure for days and sometimes weeks. Hurricane Maria’s impact on Puerto Rico is a heart-wrenching example of this: “just 15 percent of the island’s communication towers are working, and some of the island’s transmission towers have collapsed. Up to 85 percent of its fiber cables are damaged. Power remains completely out on the island, and just 25 percent of it has water service.”

So, what can utilities do to quickly rebuild an area of destruction that spans an entire island? Enter: artificial intelligence. AI applications can analyze data like equipment location and age, regional population, weather, and seasonality to identify and prioritize the highest-need areas. And when it comes to critical decisions like those that impact infrastructure, understanding the why behind an AI recommendation is essential. With this specific information, utilities can deploy drones (or their teams) with more focused objectives, improving efficiency while reducing costs of maintenance and repair.

Big Math Spotlight: Allison Clift-Jennings

Allison Clift-Jennings, co-founder and CEO of Filament, spoke with our team about the work her company is doing to scale the Internet of Things through connectivity and contractuality. She also explains why she believes blockchain technology is one of the most important contributions to computer science in the last 50 years.

By Via Science Marketing

Tell us about yourself.

My name is Allison Clift-Jennings, and though I’m currently helping to drive Filament towards this interesting world of autonomous machines, I’m an engineer at heart. I received a computer science degree and have worked within startups for the last 20 years. There’s something beautiful about the merging of new technology and entrepreneurism that I find really appealing and I can’t seem to stay away from it.

Outside of my professional life, I’m enamored with complex systems theory and like to explore it in the areas of permaculture and music composition.

Filament works to connect industrial infrastructure across secure, wireless networks building a sort of industrial Internet of Things. Can you tell us more about the Telehash, TMesh, Blockname, and Blocklet technologies Filament uses in its operations?

Filament is fundamentally a company whose core competencies involve protocol development and deployment. Most successful companies usually have some core strength that sets them apart, and ours is that our company and the people in it think very seriously and intently about how protocols can help the world operate in a more cooperative, balancing way.

Inside this area of focus, Filament has identified two components of the Internet of Things that needed significant effort to be able to scale and decentralize alongside the massive number of devices expected in the next decade. The first involves private, secure, peer-to-peer network communications — what we call connectivity. The second involves bringing economic capabilities to devices, allowing them to establish and enforce contractual agreements between themselves, other machines, and with people — we call this contractuality.

Telehash (and TMesh, its IoT device implementation)

Telehash is the protocol stack we’ve developed that focuses on the connectivity aspect. In short, it’s a peer-to-peer network protocol that allows devices to establish secure end-to-end channels between themselves and the endpoints they communicate with. Most network devices today claim to be secure, but only provide a secure link to their hub, which is often an internet-connected computing appliance. This appliance is often less secure and may use insecure network channels up to the cloud.

Telehash, in contrast, establishes the full encryption channel all the way to the cloud server it is sending its data to. This is what we mean when we say “end-to-end”. It’s been really difficult to do this in the past, because devices are not very powerful and often run off of batteries. A complete protocol was needed to allow these devices to transmit data packets across network transports such as LPWAN, Bluetooth LE, LTE/Cellular, as well as IP-based networks as used on the Internet. Telehash is a next-generation protocol inspired heavily by Jabber/XMPP* —a very popular instant messaging protocol.

Jeremie Miller, our chief scientist, is the founder of Jabber and invented XMPP and has used his experience with Jabber (currently at the height of its popularity with over 1B daily users) to inform how best to build this new network protocol for devices.

Blocklet (and its subsets Blockname, Penny Bank, and others)

Blocklet is the protocol stack we’re developing that focuses on the contractuality aspect. It’s a bit more difficult to describe Blocklet, as we don’t have many things that exist today to compare it to. In essence, this is a protocol stack that allows physical devices — the things in the Internet of Things — to extend their capabilities beyond compute and connectivity, and move into having economic capability.

Filament has a very strong ethos that claims there is raw economic value at the boundary between the physical and the digital. Specifically, the very threshold where a device can use sensors to sense physical phenomena in the real world (such as temperature, geolocation, or vibration) is where the vast majority of the value of the Internet of Things lies. Similarly, devices that use actuators to control physical systems (such as valves, switches, or control boards) also contain this vast value. Think of this capability as the reflex system of the Internet, just like the reflex system in your body or perhaps, digital skin for the Internet.

This is somewhat incongruent with the current mindset in the Internet of Things industry today that claims the majority of the value lies in the cloud. And to that position’s defense, recent developments in machine learning — specifically the recurrent and convolutional neural network developments within deep learning — has demonstrated significant value in the cloud. However, without getting valid data into the cloud, and more importantly I believe, without having the ability to effect change out into the physical world by actuation, the full potential of AI/ML in the cloud will not be realized.

To add some background on why we consider this an economic issue, consider some concepts from Economics 101. If we think about the core concepts of economics and markets, it fundamentally comes down to trust. If you buy a loaf of bread from a store, you are trusting the store to provide the bread they claim they have. You are trusting that the price of the bread won’t fluctuate significantly from one day to another. You’re also trusting that the store will accept your type of money to pay for the bread. Also, the store is trusting that you are giving them authentic, non-counterfeit money. And they’re trusting they will be able to use that money to pay for operating costs of their organization. Trust is everywhere in market economics, and to the degree that trust is unavailable, degraded, or misrepresented, a similar level of friction and inefficiency will exist.

So, in the context of Blocklet, trust is the atomic unit. If we can physically guarantee some basic levels of trust at a device level, we can then build higher-order economic capabilities on top, such as establishing value, creating markets, and even providing for devices to transact directly with each other — literally devices buying and selling goods and services between each other at machine speed. We establish this physical guarantee by using a secure element on all of our devices. This is an integrated circuit chip that gives us strong capabilities of cryptographic key storage and computation and lets us mathematically prove things like device identity (can I guarantee who I say I am?), and secure device communication (can I guarantee that what I send you hasn’t been tampered with?)

Without these protocol primitives that can ensure trust and secure communication, we believe the Internet of Things will be unable to realize its full potential.

What do industry leaders need to know about integrating this technology into their daily operations?

Contrary to the detail in which our platform was described in the previous section, it’s incredibly easy to deploy our technology in an industrial operations environment. Filament did the hard work upfront in order to make it easy to use.

A good way to describe a typical deployment is with an example. Consider a power distribution utility company. They often own and manage hundreds of thousands of power poles that provide power to their customers in urban, suburban, and rural environments. These poles are core infrastructure that have limited lifespans and they often fail from age or extreme events like heavy storms. A company could deploy our product by purchasing monitoring for their poles by simply attaching our devices to their poles. The devices are completely self-contained, including the tilt sensors, the battery, the network, and the secure element — and both the Telehash and Blocklet capabilities already embedded in it. They can deploy 1, 10, or 100,000 at once if they wish. All devices communicate with each other once configured with the proper permissions. We have other devices that can connect this large ad-hoc network to the cloud, if desired.

How did you first start researching and working with these technologies?

I’ve personally been interested in distributed systems for many years, though my professional career led me to other areas outside of this discipline. However, back around 2004, I began getting really interested in the very early work of cryptocurrencies. This was prior to Bitcoin, Ethereum, and the like. There was work being done by David Chaum with eCash, Nick Szabo with his Contracts with Bearer, and Ryan Fugger with his original Ripple protocol. For me, these were the earliest revelations that there is something amazing here if some of the core problems could be solved. One of the largest issues was this contrast between the need to trust a system, and desiring decentralization of that system, so it wouldn’t be abused later. This dilemma led to now-famous concepts like the Byzantine General’s Problem.

Our CEO Colin Gounden believes blockchain has the potential to change the landscape in the same way the Internet did. Do you agree with this estimation? How might it impact energy and heavy industry?

To continue the thought from the last section, there was a time during the mid 2000s where some interesting work was happening within distributed systems and cryptoeconomics, but it still contained fundamental problems that needed solutions. Then, there was a paper released in 2009 that described this new cryptocurrency idea called Bitcoin, and used a data structure called a Blockchain that could solve the dilemma of balancing trust with decentralization. At the time it was too early to tell if it would actually work, but in hindsight this was, in my opinion, one of the most important contributions to computer science in the last 50 years. I believe it’s up there in importance alongside spread-spectrum radio technology and the invention of the transistor. So I fully agree with Colin on this position: it’s hard to overstate how important blockchain is as it’s solved a major issue within this space. It’s interesting to consider other fields of research that are also possibly having their “blockchain” moment — most notably in the field of genetics with the invention of CRISPR/Cas9 technology.

A Harvard Business Review article published earlier this year argued that in order for a disruptive blockchain revolution “many barriers — technological, governance, organizational, and even societal — will have to fall.” What barriers to widespread adoption of blockchain have you encountered?

This is a difficult position for me to agree with. Yes, it does cause one to consider that if this technology can solve such a fundamental problem of establishing trust between parties, without authority, that it will by extension disrupt nearly all existing systems of balance and harmony that use authority. However, the Internet did just this thing 20 years ago, when it allowed people to communicate with multiple, rich modes of media, with each other, in a very decentralized way. And the television before that. And the telephone before that. And the telegraph before that. Each of those inventions did disrupt some industries — but often those industries were best positioned to take advantage of the new capabilities, though not all did.

The biggest barrier to blockchain adoption we’ve seen is simply a lack of understanding of why it’s important. Not so much how it works — because you don’t need to know how it works to gain benefit from it any more than you need to know how a telephone, or telegraph, or radio waves, or an internet router works in order to gain benefit from it. But most barriers we come across today are that people often don’t realize what blockchain-based technology can do for them. They don’t understand what problem it solves. It’s not quite as obvious at first pass as, say, a smartphone is. It’s easy to describe “computer and phone in your purse”. It’s a little harder to describe “a new form of governance and trust without coercion in your purse”. But it’s no less revolutionary.

What do you think is the next big thing to come from blockchain and IoT?

Blockchain-enabled devices is big enough!

In all seriousness, I believe we have a lot of progress to be made in getting these small devices at the edge to be much more smart — embedding machine learning capabilities directly into them, rather than running massive cloud-based data centers. While I believe there will always be a place for the cloud, it’s disproportionately distributed on the cloud side, and a movement to edge-based compute and learning could give the industry lots of new capabilities we thought were unavailable to us today.

NECPUC 2017: Recap

Last month, I was honored to be a panelist at the New England Conference of Public Utilities Commissioners (NECPUC). Thanks to our moderator, Massachusetts Commissioner Karen Charles Peterson, we had a great discussion about the opportunities and challenges facing artificial intelligence (AI) applications in regulated industries, like energy and telecommunications.

By Colin Gounden

Often AI is thought of as science fiction. But, the truth is, AI is already pervasive (well beyond speech recognition, like Apple’s Siri and Netflix’s recommendation system). In fact, you likely already use some kind of AI every day whether you are aware of it or not.

Beyond the consumer level, high-stakes organizations in regulated industries are increasingly incorporating AI into regular operations. For example, a utility company that wants to plan long-term investments in their critical infrastructure may use AI applications to determine which assets are most vulnerable.

Critical decisions, like those that affect the lives of a utility’s customers, require trust of the AI application’s predictions and recommendations. Building this trust and credibility is perhaps the biggest hurdle facing AI developers as organizations incorporate these new technologies into their operations. I recommend that developers start by prioritizing client data security. We know that AI requires data to work, but organizations are often reluctant to share their proprietary information. Providing a secure, efficient way to transfer data that serves the needs of both developers and clients is absolutely essential: enter blockchain. Blockchain has the potential to change the landscape in ways we are only starting to imagine – similar to the internet’s impact on daily life. Its ability to securely share and store vulnerable information makes it a significant innovation for the continued development of AI.

Transparency, or the “why” behind AI predictions and recommendations is another important component of building trust. Via Science works specifically to address this concern with its explainable AI systems. Most deep learning has an input and an output, but the in-between steps are a mystery. One of our key differentiators is to make the logic behind AI recommendations more transparent to people so that companies and the regulators that oversee the industry have greater trust in the decisions being made.

While developers can secure client data and provide transparency, utilities and other high-stakes, regulated organizations may still be hesitant about integrating AI. Pilots are the best way for these organizations to familiarize themselves with the potential impact of AI applications and value its potential. In fact, pilots are a growing trend across utilities in particular, as more and more consider turning to AI for modernization initiatives.

To view my full presentation from NECPUC, please visit their website.

Core Philosophy: Client Service is Essential to Successful AI Applications

Update, May 2018: VIA is now hiring for a Client Service Lead to join its Somerville-based commercial team. You can apply by emailing your resume and cover letter to jobs@viascience.com.

Last month, we welcomed Kristen Merrill, our new vice president, client service to the Via Science team. Kristen joining is the perfect opportunity to share our perspective on the importance of client service and how it applies to effectively implementing artificial intelligence (AI) initiatives (like machine learning applications).

By Via Science Marketing

Any software tool could be rendered useless without intuitive, user-friendly design and support. Via Science’s chief operating officer, Kate Ravanis, illustrates this point with an example of when our office implemented a new phone system. “Everyone knows how to use a phone, but since each system is slightly different, getting a new one means taking time to learn how to set up voicemail, add extensions, and use other functionalities. In theory, this is not hard to do. But, even taking five minutes away from your day-to-day focus to learn something new can become a source of frustration or a reason to delay adoption. In our case, the phone company sent a representative to walk us through the set-up. Of course, we could have read the manual and figured it out on our own, but having a client service representative saved us from having to even think about set-up, ultimately minimizing time lost to integrate the phones and creating a much more pleasant experience.”

Whether we’re talking about office phones or AI applications, Via Science’s core philosophy is that all new technology is adopted faster by organizations and is more effective when it comes with a strong client service wrapper. We’ve seen large organizations spend upwards of $20M on data collection efforts, only to find they lack the resources and processes to properly utilize that data. Via Science works to avoid outcomes like this; a dedicated client service expert is provided on every client engagement to make our clients’ lives easier and our applications as impactful as possible. Using some of our work in the energy industry, here are three examples of how client service is essential to building successful applications:

A client service expert helps to frame the right problem, identifies appropriate key performance indicators (KPI), and aligns stakeholders across departments.

Via Science invests its efforts in answering the right questions. Rather than simply pointing out interesting trends in data, we first determine exactly what our clients need to influence. For example, let’s say a utility company wants to reduce theft of service. Our client service expert would work to identify how this is currently done internally: what departments are involved, what data do they currently use, and what are their existing processes to help identify and reduce energy theft. Understanding these elements helps us work with our clients to vet the importance of this issue across the organization and frame the problem in a way that clearly illustrates how AI and more specifically machine learning can influence (and compare against) existing operations.

Once the problem and KPI are identified, our client service expert ensures all stakeholders in the organization are aligned around the goal and determines the roles and input needed from each team.

A client service expert determines how we will define success in both the short and long term (as opposed to waiting months or years to evaluate the outcomes of various initiatives).

Let’s return to the utility company example. Before the company can reduce theft of service, they first have to identify where it is occurring. Once they know where and how, they need to put a plan in place to reduce energy theft in the future. However, the reduction of energy theft as a benchmark for the success of machine learning applications means it could take several months to determine whether machine learning had a positive or negative impact. Even then, there are so many other factors that go into reducing energy theft (what policy changes were put in place, equipment used, budget to focus resources on theft, etc.) that separating out the specific impact machine learning made to the reduction of energy theft is challenging.

In an effort to evaluate and integrate applications more effectively, a client service expert will identify the internal metrics already in place that support the identified KPI, or as Via Science refers to them: CPM™ (Corresponding Pilot Metrics). CPM™ provide a clear way to benchmark how new machine learning applications compare to current processes and help quickly assess the potential influence using them could have on the client’s existing environment. For the utility company example, CPM™ would be the time required to develop algorithms that identify theft of service, the accuracy of theft of service instances identified, or the ease with which specific theft of service alerts can be reviewed and assessed.

A client service expert focuses on how clients will use the application in their current environment. The expert relays this information to Via Science’s development team to ensure the final product we build can be seamlessly integrated into the client’s existing workflow.

In order to make our clients’ lives easier (and provide them the most value) we strive to create applications that fit into their current operations. As we mentioned earlier, client service experts seek first to understand how the identified problem is currently solved within the organization (which teams, systems, and data are currently used). Awareness around these elements is critical to ensuring we build an application that can easily be adopted and integrated into existing workflows. For example, we would look to build a first version of the application using data that is already centralized or potentially already used for detecting theft of service purposes. This will ensure faster adoption and easier integration than if we created a system that requires new data be collected across multiple stakeholders (who may not have been involved before).

We understand that needs and expectations can change throughout the course of engagements as we learn more about the problem. Our client service team works to proactively seek out feedback (specific requests, unexpected events, changes of focus) to continue refining the application to better meet client needs.

There is no shortage of content boasting the potential of artificial intelligence applications to make organizations more efficient, effective, or successful. However, applications must address critical challenges and fit within an organization’s workflow in order to reach this potential. Via Science believes its client service team supports the success of both its applications and its clients by communicating key information across key stakeholders. We are passionate about making sure the tools we build are highly customized, user-friendly, and flexible to grow and change with evolving client needs.