The Kore is our automated ML platform, the engine that drives AI. We created it to rapidly build, optimise and deploy AI models in order to accelerate the data science process. What before would take months The Kore is able to deliver in days. The Kore automates every step so businesses can generate value faster.
The potential of AI and Machine Learning in business is well articulated. But the problem for many businesses is knowing where to start. Our four week Accelerated Intelligence programme is designed to deliver a practical application of AI in just one month.
Using The Kore's powerful Auto ML, Data scientists can focus on feature engineering rather than trial and error aspects of model selection and tuning. Non-experts can learn how to unlock the power of Artificial Intelligence and quickly apply it to their business problems in minutes.
A web interface allows fast deployment of trained models to scalable robust APIs which allow simple integration into existing systems and processes.
Uses AI to find and tune the best model to solve any problem.
Exploit parallel computing to find optimal solutions in minutes not months.
Productionise models to meet enterprise level SLAs with a few simple clicks. Use models via UI or API.
Published models can handle tens of millions of real time customers.
Support models in beta, production, etc. seamlessly, including management, audit, rollback, testing and more.
Audit the decision making of any trained classifier.
“The effects of AI will be magnified in the coming decade as industry transforms their core processes and business models to take advantage of machine learning. The bottleneck now is in management, implementation, and business imagination.”
Erik BrynJolfson & Andrew McAfee - Harvard Business Review
We run a half day workshop with the relevant stakeholders. There are three key questions we work through - how do you make money, where do you lose money, and what data do you have available? From this starting point, we generate potential uses for AI in your business. The aim at this stage is to go broad and uncover as many opportunities as possible.
Each use case is assessed for viability. We look at level of effort required, speed of adoption/implementation, quality of available data etc. With your team we look at existing software, processes and structures and collaborate to understand the potential benefit of each use case. From this process, we can identify the key uses cases to consider for pilot.
We take a sample of the data and create a proof of concept. This PoC demonstrates the viability of the use case, showing how actionable, impactful results can be achieved from the application of AI. We use our Automated Machine Learning platform, The Kore, to accelerate the data science process so real results are available in days.
Drawing on insight from Weeks 2 and 3, we define a high level rollout plan, covering required processes, data, resources and timings to give a clear idea of how the Proof of Concept can be integrated into your business. We will leave you with a functional prototype producing real business results, and the tools needed to build out greater capability.
Kortical are collaborating with a major UK retail bank to enter the era of cognitive banking.
After examining the business and market, we defined over 70 possible use cases. We used The Kore to deliver five Proof of Value projects in less than three months. Two are currently entering full production:
Predicting customers future financial distress with 93% accuracy and optimising automated customer interventions using a next best action system.
Combining RPA, NLP and The Kore to massively increase efficiency in processing customer emails - automating a £1m cost center. The Kore allowed us to move from data to deployed service in under a day.
Ecodesk.com is a live web-based platform which enables organisations of all sizes to search, publish, analyse and communicate sustainability data in one place. They work with 12,000+ companies in over 100 countries.
Currently businesses use Ecodesk to engage with and report on their supply chain. This has built a strong database of suppliers and associated sustainability data.
We used The Kore to unlock the predictive power of their data. Supplier spend and emissions can be predicted. Supply chains can be optimised up front to minimise spend and environmental impact.
Ecodesk’s service moved from reporting on a cost and compliance issue to actively reducing and managing environmental impact.
A major investment bank runs over 5,000,000 trades each night, on 80,000 computers, to calculate the risk on each position. Humans were manually setting parameters to try to group the trades so they could be calculated efficiently. But there were 400 parameters with over over 12,000 combinations. This problem is entirely intractable for humans.
We added extra instrumentation to the Risk Platform to track all the metrics we’d need. We then used The Kore to create a model that could predict how long different trades would take to calculate. Using this we were able to target a desired runtime for each job that would use resources much more efficiently. Finally we integrated this into their trade batching workflow.
They saw an improvement of 30% in their overall runtime after just three days. The platform continues to learn and adjust its model to refine the improvements and adapt to further changes.
Alex has a degree in Artificial Intelligence and a PhD in Data Science and, prior to Kortical, imagined, designed and deployed AI systems in enterprise as independent consultant. He specialises in understanding both business need and AI capability - able to bring both together to identify high leverage oppertunities.
Andy specialises in high performance computing, winning awards from NVIDIA for his work in optimising quantitative financial models on CUDA enabled GPUS. Prior to Kortical, Andy has led successful machine learning initiatives within a both a Tier 1 investment bank and financial software developer.
Former U.K. head of strategy at HUGE, Martin has led cross-functional all-agency teams, on projects ranging from marketing to product development and the vast grey area between, on a local and global level. Martin is Marketing director at Kortical.
Formerly European Head of Ecommerce & B2B Digital Marketing at Coca-Cola, Barbara's background was in agencies such as Elvis.
Barbara is head of commercial operations at Kortical.