Embrace Home Loans, a prominent mortgage lender licensed in all 50 states and the District of Columbia, sought to optimize its marketing spend across its digital and direct mail channels. The company wanted to maximize marketing spend and increase revenue across all marketing channels. The challenge was to do so across the scale of Embrace’s operations, which was a significant task. The company needed a solution that could manage hundreds of Jupyter notebooks and run SQL queries on millions of rows of data. The solution also needed to ensure the security of Embrace’s customer data, which included risk-based and standards-based security protocols to protect all data.

Citi Ventures, the innovation arm of Citibank, is constantly on the lookout for emerging trends in technology and financial services that can help solve challenges faced by Citi and its clients. Since its inception in 2010, Citi Ventures has invested in over 100 different companies to enhance Citi’s products and services. However, the organization was seeking innovations that could solve challenges for Citi and its customers more efficiently. They were particularly interested in the field of AI and machine learning, which they saw as game-changing for the financial industry. They were looking for a solution that could empower both data scientists and business users, automating much of the modeling process and freeing up their time to focus on solving complex business problems.

Sanlam, Africa’s largest non-banking financial institution, exists with the purpose of empowering generations to be financially secure, prosperous, and confident. However, the company was facing challenges with its data science operations. The open-source AI options they were using felt cumbersome to navigate and lacked critical explainability for business stakeholders and compliance. This was hindering their ability to drive critical business value levers such as sales and client retention. The company needed a more streamlined and transparent AI solution that could help them improve their operations and deliver better results.

Freddie Mac, a company chartered by Congress in 1970 to support the U.S. housing finance system, has been facing challenges in achieving meaningful predictions and key insights to inform business decisions. The company works with hundreds of thousands of customers and mines nearly four terabytes of data. However, they found that business intelligence and manual practices didn't scale effectively across this vast customer base and data volume. As market and economic conditions change, Freddie Mac must remain flexible and continuously deliver on its commitment to affordable, adequate housing. In a sea of unstructured and semistructured data, it’s challenging to achieve meaningful predictions and key insights to inform business decisions.

Cegid, a French tech company offering cloud services and management software solutions, is facing the challenge of creating more models in less time while minimizing the technical skills and resources required. The company serves 350,000 customers across 150 countries and generates €632 in revenue. The predictive analytics team at Cegid is under pressure to meet the ever-expanding demand fueled by frequent acquisitions. The team is tasked with tackling a growing list of business challenges, including predicting the likelihood of getting paid on invoices and the propensity of customers to add services.

MAPFRE, a Spanish insurance company, operates in over 100 countries, generating €27.3 billion annually. The company's analytics team is responsible for providing advanced analytics to help make decisions on pricing, sales, retention, underwriting, and more. However, given the demand for data insights, the team found it challenging to keep pace with the many incoming requests and deliver value quickly. The team needed to expedite its time to market in tackling new business challenges.

AUTOproff, a European leader in digital dealer-to-dealer trading, was facing a challenge in scaling its operations. The company, which had more than 100,000 cars on auction in 2021, was struggling to produce car value estimates within the 20 minutes promised to customers. This task was entirely dependent on a team of skilled vehicle professionals. As the company grew, the need for scaling became increasingly important. The challenge was to automate the process of producing car value estimates to expedite the turnaround time for customers and free up the data scientists and estimators to focus on more rewarding parts of their jobs.

Decode Health, a healthcare AI company, has always relied on predictive analytics to unlock discoveries using data. However, in the early days, modeling was a slow, manual task. Analyzing a single dataset could take two to three weeks, with two to three data team members working around the clock. This exhaustive manual effort included considerable time preparing data, waiting on models, recalibrating, and waiting again. The company needed a solution that could streamline this process and deliver accurate results more quickly and cost-effectively.

Phoenix Children’s is one of the nation’s largest pediatric health systems. It provides world-class inpatient, outpatient, trauma, emergency, and urgent care to children and families for more than 38 years. The organization is continuously at the forefront of innovation and is recognized among the nation’s top-ranked children’s hospitals. Phoenix Children’s wanted to use analytics to improve both clinical and operational decisions. However, manually building a single model took the better part of a year. The healthcare system knew that a certain percentage of children who present with other health concerns may actually have undiagnosed malnutrition. If they could identify cases of malnutrition, they could intervene and influence outcomes.

Carbon Digital Bank, a financial institution serving the underserved African market, needed a way to quickly determine credit risk for individuals without prior credit. The bank also wanted to empower its data science team to take on additional business challenges. The bank had committed to a data-first strategy and looked to AI as an integral part of its decision-making. However, assessing customers' credit worthiness was a major challenge. The bank needed to expedite decisions on hundreds of thousands of loan applications every month.

A leading US-based fintech company with a development center in India was facing difficulties in monitoring process compliance across its numerous ongoing projects. The company lacked centralized visibility to assess compliance across enterprise projects. Manual tracking of every commit, pull request (PR), and peer approval was untenable. It was also challenging to track if developers used the predefined tools and procedures for version control, source code management, peer reviews, etc.

The company, a leading media and entertainment entity with a presence in over 150 countries, was facing challenges in managing its applications, including a newly launched subscription-based streaming application. The company's internal DevOps team was responsible for managing these applications, but the company wanted to improve visibility into performance, identify areas for improvement, and gauge customer experience. However, they lacked a standard framework to measure DevOps success and relied on monthly manual reports to understand the team's health and performance. This approach had limitations in analyzing DevOps data and metrics. Furthermore, frequent bugs and a longer time to resolve issues led to a poor customer experience.

A leading U.S.-based wireless and telecommunications service provider wanted to improve call center performance, increase customer satisfaction, and have greater insight into the activities of its call center representatives. To achieve this, the Fortune 50 Company wanted to analyze the desktop activities of the call center representatives around the clock. The client wanted to monitor desktop activities in real-time while the representatives are on duty. From an operational perspective, this meant creating a centralized system where operations personnel would be able to track idle time, track what websites are being used for how much time, track outlook usage, and track various applications being used on the desktop. The client also wanted to track desktop activities when the agent are on call, not on call, and on call and kept customer on hold.

The customer, a US-based Fortune 100 broadband connectivity company and cable operator serving more than 30 million customers, was facing several technical and business challenges with their existing data workflow. They received data from multiple sources that was fed into an SFTP server. After ETL was performed, the data was read by an Informatica workload and persisted to their Teradata data warehouse. Business analysts then accessed this data and ran queries to gather insights. The client wanted to make a strategic shift to the cloud to enhance scalability, reduce costs, improve query performance, realize a unified view, simplify management, seamlessly integrate with other cloud-native services, and automate workflows for CI/CD.

A leading media and entertainment company with a presence in over 150 countries and a headcount of over 3000 employees faced several challenges in managing their applications. They had recently launched a subscription-based streaming application, in addition to their existing apps that required frequent updates. Their internal DevOps team was responsible for managing these applications, but the company wanted to improve visibility into performance, identify areas for improvement, optimize costs, and assess customer experience. They lacked a standard framework to measure DevOps success and relied on monthly manual reports to understand the health and performance of the team. They also faced limitations in analyzing the DevOps data and metrics. Frequent bugs and a longer time to resolve issues led to a poor customer experience.

The client, a major telecom company providing nationwide telecom services, was in need of a system that could perform real-time, multi-lingual classification and sentiment analysis of text data. They were looking for a solution that allows storing, indexing, and querying PetaBytes (PBs) of data with a very high throughput. The critical requirements included the ability to ingest and parse a high volume of data [250M (15 TB) records/day] of varied types such as weblogs, email, chat, and files. They also needed to apply real-time multi-lingual classification and sentiment analysis with very high accuracy (four nines), store metadata and raw binary data for querying, and meet a Query SLA of 5s on cold data.

Ecolab, a global leader in water, hygiene, and infection prevention solutions, wanted to develop predictive risk models for water systems, industrial machinery, and other applications. The company's machine learning journey began in 2016 with a project to develop bacterial growth risk models using existing sensor data. However, the process of building, deploying, and maintaining machine learning models in production was complex and challenging. The company needed a data science collaboration platform that would bring together its large, geographically dispersed team, while efficiently using cloud computing resources. The deployment of machine learning models at Ecolab followed a 'rewrite-and-deploy' pattern, where model development occurred independent of the application developers. This approach led to deployment timelines exceeding 12 months on average.

Researchers at EMBL, Europe’s flagship laboratory for the life sciences, were looking to enhance traditional microbiology methods with Deep Learning. Their goal was to reconstruct the complex biological phenomena that underpin the life cycle of cells. This was a significant challenge due to the complexity of cell life cycles and the limitations of traditional microbiology methods. EMBL operates across six sites in Europe and has more than 80 independent research groups covering the spectrum of molecular biology. The challenge was to develop a solution that could accurately model the lifecycle of cells and provide insights into complex biological processes.

Epona Science is a company that specializes in buying, breeding, and identifying the best racehorses in the world. The racehorse business is a traditional industry where buyers often rely on pedigree or trusted breeders' instincts to choose horses. However, Epona Science believes that these are not the best predictors of success. They aim to revolutionize the industry by using machine learning, statistical analysis, and science. They have discovered that factors such as the horse's entire genetic profile and lineage, its height and gait, and even the size of its heart can make a significant difference in its performance. However, gathering all this data, cleaning it, standardizing it, and getting it into a consistent format that their machine learning models can train on is a significant challenge. The data comes from various sources worldwide, including x-rays, genetic profiles, and track records from previous races.

RTL Nederlands, part of Europe’s largest broadcast group, wanted to use artificial intelligence (AI) to make video content more valuable and discoverable for millions of subscribers. The company broadcasts to millions of daily TV viewers, along with delivering streaming content that garners hundreds of millions of monthly views online. One of the key growth metrics for RTL Nederlands is viewership, but optimizing the value and discoverability of video assets is an extremely labor-intensive endeavor. That makes it ripe for automation, and the team applied machine learning to optimize key aspects of its video platform, like creating thumbnails and trailers, picking the right thumbnail for those trailers, and inserting ad content into video streams.

One of the top for-profit managed healthcare providers in the U.S., with affiliate plans covering one in eight Americans for medical care, was looking to leverage artificial intelligence (AI) to harvest long-term insights and make much more detailed health predictions from claims and electronic health record data. The data store is massive, with more than 50 terabytes of data covering the company’s tens of millions of members across the U.S. They were mining this data to determine treatment efficacy based on past outcomes given particular patient characteristics. However, getting these potential insights into the hands of healthcare providers was a challenge. It’s one thing to have small scale implementations working in a lab, it’s another to deliver machine learning at scale. When the engineering lead joined the AI team, they had a very complicated data delivery pipeline based on Apache Airflow. While it worked, it wouldn’t scale beyond a single pipeline or container instance at a time.

Adarga is an AI software development company that provides organizations with the capability to build and maintain a dynamic intelligence picture. Its AI analytics platform processes huge volumes of unstructured data, such as reports, global news feeds, presentations, videos, audio files, etc., at a speed unachievable by humans alone. The software extracts the essential facts in context and presents them in a comprehensible manner to unlock actionable insights at speed and enable more confident decision-making. However, the company faced challenges in developing, training, productionalizing, and scaling the necessary data models. They needed a solution that could drive data consistency, understand lineage, and enable model scaling.

SeerAI’s flagship offering, Geodesic, is the world’s first decentralized platform optimized for deriving insights and analytics from planetary-scale spatiotemporal data. Working with spatiotemporal data is a challenge. Because it concerns planetwide questions, the data sets are massive in scale – often entailing petabytes of imagery. The data itself can come from different sources, requiring the ability to load and manage from a decentralized data model. Finally, that data is generally heterogeneous and unstructured, and thus notoriously complex and difficult to deal with. SeerAI designed Geodesic to constantly grow in knowledge and data relationships so that it can eventually answer most any question. Controlling the data ingest, ML job scheduling, model interaction, and data versioning can be extremely complex at this scale.

Riskthinking.AI, a company specializing in measuring the financial risk of climate change, was in the early phases of ramping up their internal AI infrastructure when they took on the CovidWisdom project. The project was a response to a call from the Canadian government to assess the economic impact of major pandemic policies. The challenge was to predict the best way to implement societal-level responses like lockdowns with the minimum amount of damage to daily life and the economy. However, the team realized they had experts in predicting the future but not in building AI architecture. They had data scientists working on laptops, pulling and pushing data over VPNs to remote work spots, and even building their own Docker containers. They needed to move from ad hoc to MLOps.

Wayve, a London-based company developing artificial intelligence software for self-driving cars, was facing a significant challenge with their GPU resources. Their Fleet Learning Loop, a continuous cycle of data collection, curation, training of models, re-simulation, and licensing models before deployment into the fleet, was consuming a large amount of GPU resources. However, despite nearly 100 percent of GPU resources being allocated to researchers, less than 45 percent of resources were utilized. This was due to the fact that GPUs were statically assigned to researchers, meaning when researchers were not using their assigned GPUs others could not access them. This created the illusion that GPUs for model training were at capacity even as many GPUs sat idle.

The company, a world leader in facial recognition technologies, was facing several challenges with their GPU utilization. They were unable to successfully share resources across teams and projects due to static allocation of GPU resources, which led to bottlenecks and inaccessible infrastructure. The lack of visibility and management of available resources was slowing down their jobs. Despite the low utilization of existing hardware, visibility issues and bottlenecks made it seem like additional hardware was necessary, leading to increased costs. The company was considering an additional GPU investment with a planned hardware purchase cost of over $1 million dollars.

The London Medical Imaging & AI Centre for Value Based Healthcare was facing several challenges with its AI hardware. The total GPU utilization was below 30%, with significant idle periods for some GPUs despite demand from researchers. The system was overloaded on multiple occasions where more GPUs were needed for running jobs than were available. Poor visibility and scheduling led to delays and waste, with bigger experiments requiring a large number of GPUs sometimes unable to begin because smaller jobs using only a few GPUs were blocking them out of their resource requirements.

Exscientia plc is an AI-powered drug discovery organization that relies heavily on the accuracy and stability of its models. The company's model deployment process is unique as it is entirely automated, resulting in thousands of models being delivered, monitored, and retrained without human interaction. However, as Exscientia expanded its reach and goals, it needed an enterprise-grade scale solution. The team was looking for additional operational efficiencies and other ways to debug and stabilize models. The existing open-source deployment solution and inference platform were no longer sufficient for their growing needs.

Capital One, a leading US retail bank, was facing significant delays in their machine learning (ML) deployment pipeline. The data science teams were heavily reliant on the engineering department to test, deploy, or upgrade models. This resulted in month-long lag times and the need to redeploy entire applications for updates to existing models. Scaling up projects was only possible by using more developer resources and people power, which further strained the already overstretched teams. The bank needed a robust, scalable, and flexible approach to the deployment of ML models to support its millions of customers and users of their mobile banking app.

Noitso, a company based in Copenhagen, Denmark, specializes in data science, data collection, and predictive analysis. They provide their customers with credit ratings, scorecards, and risk profiles using data science and AI. However, they faced challenges in deploying their models. The models took a long time to get to production and lacked explainability and monitoring. They were unable to determine when models needed to be retrained, and had to do it after a fixed period of time rather than when necessary. This approach was the only way to maintain accurate predictions and prevent issues such as data drift.