AWS for Industries

Humans as Sensors for Water

Blog contributed by Mayank Sharma, Chief Data Officer at Vyntelligence.

Humans have a vital relationship with water – it is essential for life, it nourishes and delights us, and it has been and continues to be revered as a fundamental element of nature. In fact, our thirst response mechanism is based on a highly sensitive chemical threshold, making us sophisticated water sensors in a very literal sense.

Image by Vyntelligence, an AWS Energy Competency Partner

And yet, when it comes to addressing the challenges we face in building environmentally friendly and sustainable societies, our role as citizens of the world has been largely confined to that of a consumer or a spectator. The technological response to these problems has so far focused primarily on wide-scale instrumentation and automation.

We have all heard the forecasts that suggest that by 2030, around 50 billion sensors and devices will be connected to the cloud. But with all the focus on Internet of Things (IoT), artificial intelligence and machine learning (AI/ML), and other technologies, are we perhaps missing the opportunity to harness humans as sensors in this endeavor within the water industry and more generally? After all, the Human-as-a-Sensor (HaaS) paradigm posits that there are billions of mobile phone users across the globe, equipped with functionality-rich smart devices and contextualized situational awareness that can be enlisted as participants in this grand initiative.

A couple of obvious areas where the use of humans as sensors could be a force multiplier are in

  1. Education: getting people to understand the value of water
  2. Discovery: capturing rich multimodal data

By empowering customers to better understand the value of water by having them participate in a HaaS construct, we uplift their ability to appreciate and articulate the challenges in delivering water, treating wastewater, as well as the overall value of the entire ecosystem of water services. And if we can allow people to collect the right data at the right place and at the right time using well-designed tools, we vastly improve the ability of AI/ML to deliver utilities insights on and visibility into the state of the water system.

Vyntelligence Solution on AWS

Vyntelligence provides a smart-video based digital work intelligence platform that facilitates collaboration and AI-assisted decision making for analog, people-centric processes. Through prompted data capture and guided workflows, Vyntelligence ensures that customers and field workers can easily collect the right data at the right time. These multi-modal inputs – which include videos, images, bar codes, and form data – are analyzed to extract hard-to-get insights using industry-leading machine learning models. These insights are then channeled to all entities through their collaboration layer to support rich shared analysis and decision-making steps such as annotations, audits, delegations, notifications, and approvals. This allows an enterprise to empower its customers and employees and hence achieve higher customer satisfaction or net promoter scores, better first-time right metrics, and find operational efficiencies through knowledge management and sharing.

Developed and hosted on AWS infrastructure, Vyntelligence takes advantage of many AWS services – Amazon S3 for efficiently storing and indexing large amounts of video data, reliable scalable and secure web and database provision using Amazon EC2, Amazon Relational Database Service (Amazon RDS).

Our proprietary AI/ML models that are used to process video data can be scaled horizontally Amazon Elastic Container Service (Amazon ECS) to bring down completion times for urgent workloads. Content delivery is achieved securely with high speed and low latency using Amazon CloudFront and AWS Web Application Firewall.

AWS support for DevOps, software deployment, infrastructure as code, role and policy-based least privilege access control allows Vyntelligence to meet the high operational and data security compliance requirements of even the most demanding large enterprise applications. Securing access and auditing our process is accomplished with AWS Identity and Access Management (IAM). Encryption keys and other sensitive authentication tokens are rotated and kept safe through AWS Key Management Service (AWS KMS) and AWS Certificate Manager; including our dedicated hardware based key material.

Northumbrian Water Limited Case Study

Northumbrian Water Limited (NWL), a UK company, wanted to empower customers to perform self-assessments through digital field service forms. This would provide a highly personalized customer experience and support virtual collaboration between staff, third-party contractors, and customers.

NWL worked with Vyntelligence to deploy a video-based self assessment and remote auditing tool using Vyn SmartVideoNotes. Customers now report issues, such as water leaks, to a service provider’s call center using a short guided video captured on any mobile device – even without a signal. The customer’s self-assessment video is guided by an automated storyboard that prompts them to film the key data points required, such as “the origin of the leak.”

The AI-powered platform automatically labels the short video and integrates it into the customer service workflow to be audited remotely by call center agents and then assigned to the appropriate technician. Vyn SmartVideoNotes provides a 360-degree view of each customer’s issue without the need for an initial callout. The severity of each case is assessed remotely, and rich data insights with predictive analytics recommend next best actions.


Vyn SmartVideoNotes was rolled out within three weeks to achieve business continuity for the NWL during COVID-19 while keeping workers safe and delivering a superior customer experience. Vyn SmartVideoNotes saw a strong uptake as 62% of customers agreed to record self-assessment videos, reducing trouble call duration, empowering customers, and increasing customer satisfaction.  The number of field crew visits at the initial investigation stage were reduced by 32%, significantly cutting operational overheads. Using Vyn to digitize field service forms likewise reduced case handovers by half, improving workforce productivity and customer response times as a result.

Figure One: Vyntelligence Results at Northumbrian Water Limited

Nigel Watson, CIO Northumbrian Water Group said, “Video seemed like the next logical step. We started with health and safety and that worked really well. We were looking at ways of reducing the need to go into customer homes. We decided to make it available to customers for reporting leaks and floods. We found during the pandemic that with Vyn we reduced field visits by 32%. The best feedback we heard was from the field staff that they felt cared for.”

To learn more visit Vyntelligence or the Vyntelligence AWS Marketplace page.

Mayank Sharma

Mayank Sharma

Mayank Sharma is Chief Data Officer at Vyntelligence, the category-defining Smart Video Notes platform for enterprise work assurance. In this role, he is responsible for driving product and solutions strategy with a single-minded focus on helping customers realize their business outcomes around quality, efficiency, safety and sustainability through data-driven insights. He is an experienced technologist and thought-leader with a long track record of business and scientific impact, particularly through the use of process optimization and AI to solve complex business problems. During his career, first as a Research Staff Member at the IBM TJ Watson Research Center and then as Head of Data Science and Vice President of Technology at Raymond James, he has been at the forefront of innovation in diverse areas such as human capital management, business process automation, healthcare analytics, smart grids, FinTech and EdTech. Mayank obtained a B.Tech. from IIT Delhi; and an M.S. and Ph.D. from Stanford University, all in Electrical Engineering. He has a broad background and expertise in data analytics, operations research, machine learning and process redesign. His work has been recognized through various IBM internal awards; the 2010 Daniel Wagner Prize for Excellence in Operations Research Practice and the 2019 Applied Probability Society Best paper prize, both awarded by INFORMS. He is Member of the Forbes Technology Council, a Senior Member of the IEEE, and a member of INFORMS and ACM. He is widely published in the scientific literature; has been granted 19 patents; and has been an Adjunct Assistant Professor with the department of Computer Science at Columbia University.