Spraoi: Enabling Insurers Leverage Cutting-edge Machine Learning Solutions

Follow Spraoi on :

Spraoi

Karan Mishra and Matthew O’Mara and Santoash Rajaram , SpraoiKaran Mishra and Matthew O’Mara and Santoash Rajaram
As a company that has mastered building solutions that leverage troves of carrier data to generate valuable insights, Spraoi has indeed established a unique stance for itself in the AI technology space. For Spraoi, it begins with the establishment of trust.

Carriers consider data to be the most important asset, and driver of both growth (identification of new markets, development of more targeted products/services) and optimization (more efficient routing of service and claims, identification of fraud) opportunities. Spraoi’s initial clients showed a great deal of trust in sharing their data, and it wasn’t a covenant that the company took lightly. To this end, the company invested its time and energy beyond contractual obligations to not only deliver machine learning solutions but also educate as to how improve the quality of their data (e.g. recommendations for improvement) and other solutions that could be leveraged. Many consider machine learning solutions to be a black box. Spraoi believes that better-informed customers can expand the aperture of how these solutions could benefit them now and in the long term. Accordingly, the company implemented privacy and security protocols that showed clients that their trust was well placed.

Spraoi, since then, has expanded its reputation in the marketplace by introducing innovations based on their learnings delivering production machine learning solutions. By applying the lessons learned from 30 months of delivering machine learning solutions, Spraoi has built its BARREL platform and SaaS models to address the various predicaments of insurers as it provides an infrastructure to create self-describing datasets with built-in error handling.

Following is the conversation that CIOApplications had with Karan Mishra, the co-founder of Spraoi, where he narrates how the company brings use cases and proofs of concept to facilitate ideation, and assist clients who don’t have historical data to build a solution. He also talks about how Spraoi guides its clients through an evolutionary journey, operationalizes its models with options (direct API or leverage of our user experience), and provides enhanced security features (hosting options and filtering PII/PHI).

What are the market pain points? How does Spraoi improve the purchase, service, and claims processes for stakeholders? Please elaborate.

We have worked with clients in several insurance industry sub-segments. We have developed a claims suite addressing segmentation, assignment, auto-adjudication, and fraud detections, with LOB-specific models (for disability, we have built propensity to settle, identify offsets, and understand what short term disability claims are likely to become long term claims). We have built models in the retail life and annuities space such as propensity to purchase, application fraud, distribution fraud, employee fraud, and policyholder propensity for lapse/recession.

How does the BARREL platform’s patent-pending infrastructure afford Insurance specific, reusable rules, schemas, and processes to speed the machine learning model development process and virtually eliminate DevOps and Data Engineers from it?

With our platform, we have addressed the impediments to and accelerated the delivery of machine learning models for insurance carriers. Our production solutions have proven this in the market. Barrel leverages insurance specific, reusable rules, schemas, and processes to speed the machine learning model development process and virtually eliminate DevOps and Data Engineers from it.

The Spraoi team has matured BARREL over nearly three years as an insurance-specific machine learning delivery solution

Barrel offers a simple way to assemble workflows using reusable operations. Operations, once defined, can be used against any self-describing dataset making workflow simple to configure and use. It can provision dashboards to specific users and pre-built operations in workflows to keep the dashboards up to date. Further, Barrel combines the experiment infrastructure with the production workflows enabling the data science workflow to span the entire data supply chain.

The Spraoi team has matured BARREL over nearly three years of insurance-specific machine learning delivery. We are now in a position to go beyond SaaS models to BARREL’s license as a standalone platform for clients’ data science teams. Should clients still wish to receive support, we are willing to do so on a consultative basis.

How does Spraoi make sure that its offerings help organizations to understand real-world requirements and fulfill them?

Our process and platform yield results. Spraoi’s team of analysts, data scientists, and engineers develop SaaS machine learning solutions that can be delivered via API or operationalizing user experience within 12-24 weeks. Our evergreen process integrates stage gates to ensure that we have what we need and enable transparency to both the process and impact of each portion of the process on the whole, which includes:

• UNDERSTANDING the why of the problem, consider what data is available, drafting and iteratively adjusting the model(s) to fit the data best,

• DESIGNING an experience for the intended user group, and considering desired confidence thresholds and adapt accordingly.

What impact does Spraoi’s work culture have on its employees and clients?

The culture of our firm started with the naming. Spraoi is the Gaelic word for fun. The founding team has emphasized the importance of enjoying the process of delivering high-quality products and services for our clients. The team is a reflection of our founders, the combination of technology, and consulting capabilities. While we all work hard, we engage in regular interactions that go beyond work topics. As a geographically diverse team in the US and India, we get together for “social hours” to engage in discussions on topics of the day, play games (Pictionary and Jeopardy! have been favorites), and share movie/TV recommendations. It makes us closer as a team and the tightness of the unit has paid dividends. We have extraordinary employee retention. We bring multi-disciplinary teams to our sales and delivery interactions. Clients feel the difference and see the difference (e.g., tight-knit teams that generate quick turnarounds). They like working with us as much as they like the output of our teams.

During client delivery, our tightly integrated teams and approachable, consultative demeanor puts clients at ease and fosters additional requests. We were originally engaged to work on a solution integrating data from the client’s recommended optical character recognition (OCR) vendor. The OCR vendor was not up to the task. At the client’s request, we took on the responsibility to drive the identification, incorporation, and ongoing management of the solution's OCR portion on the client’s behalf. We aren’t just a rigid product provider. Our consulting heritage has engendered a “can do” attitude that our clients like and respect.

Can you provide us a sneak-peak as to what lies ahead for Spraoi?

Having worked to deliver production machine learning solutions in the marketplace, we realized it was difficult to credibly scale Data Science capability in a profitable way. We have responded to the needs of carrier clients by developing BARREL, our patent-pending machine learning platform and associated SaaS models. But we are not complacent. If we have learned anything over the last three years, it is to be constantly advancing. To that end, we are working on three capabilities clients will find valuable:

• ML-POWERED SCHEMA INFERENCE. For datasets that are ingested into the system, we will leverage machine learning to help create self-describing datasets with no manual intervention.

• BUSINESS USER WORKFLOWS. Integrating human workflows that can be combined with system workflows to enable business users to be part of the ML process and more tightly integrate models outputs and business outcomes.

• RECORD LEVEL LINEAGE TRACKING. Every record that goes through Barrel will be tracked as will the lineage of how it traveled through the system. While we currently do this at a dataset (file) level, we will now surface it in the user experience.

Read Also



Featured Vendors