Building a bespoke machine-learning app in a single day sounds improbable, but it is, in fact, now possible. Just ask the Founders of award-winning AI startup Kortical, who showed a new app to the FCA that predicts pensioner life and financial satisfaction 20 years ahead of time, with >80% accuracy. The app in question can visualise individual spending habits and even text them to intervene. Amazingly, it was conceived, automatically coded, and performing within a single day; a child app of Kortical’s primary and proprietary AI technology: the Kortical platform.
What Kortical has created is remarkable; a powerful Auto ML toolkit that can quickly create any number of bespoke new analytical apps, on-demand. The primary AI responds to extremely complex data sets (i.e. genetics), and the direction of data scientists, automating much of the process, while giving data scientists control over the models and deployment. Together, they can make invaluable new digital tools, specific to a client’s needs; from optimising kidney donor matches or analysing hospital maintenance schedules (both real examples). Machine Learning software creation has become semi-automated, using a patent-pending blend of ‘Auto Machine Learning’ (Auto ML) and data-scientist know-how.
It promises clients like big banks, big consultancies, or indeed little startups, the opportunity to do extraordinary analyses of complex data, with bespoke tools, without hiring an army of data scientists, or indeed hiring any AI software developers. Rather than eliminating data scientists, the company sees itself assisting them to build much better models. “New models can go from prototype to enterprise-scale deployment in a few clicks,” according to COO Barbara Johnson.
Recruiting top data-scientists is tough, and expensive – ditto for data engineers, so Kortical believes it can help clients to close skill gaps, and save money, by enabling their team to do much more. Kortical’s business model is SAAS, which promises a substantial acceleration in AI technology, at lower costs.
NHS Blood and Transplant services are an example of how Kortical has been used to cut costs and wastage; by as much as 50%. For this project, Kortical was able to prove the benefits within just 6 weeks from receiving raw data. Johnson explains how, on a blood supply project, the NHS needed to understand day-to-day supply metrics in greater depth, and get clarity from it. A bespoke app was created for this analysis, processing extraordinary volumes of corresponding data, including the effect of blood supply during school holidays, or major sports events. The Blood Service can now predict supply & demand in real-time; a leap forward in risk management, a significant increase in efficiencies, and no doubt, saving lives too.
Impressively, any new models created by the primary AI are explainable to key stakeholders. This can be very important for GDPR, for example, for auditing, or for commissioning executives who need to present their work confidently, to justify the strategy and further investment. Automated machine learning like this is now enabling businesses to achieve in days what once took months, or was impossible altogether.
Kortical Startup Story
Building a new AI platform to automate data science isn’t easy, neither is securing clients like EE or Deloitte. Both co-founders are technical experts, with deep expertise in this domain; CTO Alex Allan has a PhD in AI and a background in commercial Data Science, while CEO Andy Gray is similarly a software engineer, with a background in AI, having worked with Barclays Capital, and previously as lead AI/gameplay programmer at EA games.
The founders were able to partner with a big consultancy for launch, prior to raising capital, and test the technology with blue-chip clients. Kortical now offers clients a monthly subscription with data science support. Bespoke apps created for clients can be regularly updated, while costs are spread out through subscription.
First-round investment involved a £1.2million seed round, funded by a group of tech Founders as angels, and 2 corporate investors. Kortical chose SeedLegals as an innovative LegalTech platform to run the funding process through, over a conventional law firm, having shopped around and sense checked best practice.
The legal process included COO Johnson generating TermSheets online, enabled by the platform. It brings transparency and simplicity to the creation of investment documents, which are all created on the cloud. Investors can log in remotely to review Term Sheets, and sign digitally when they are agreed. Finalising the completion of a round has been made much easier, as new Articles and an Investment Agreement documents are automatically created, alongside relevant filings with HMRC and Companies House.
Johnson loves the efficiency of the platform, and how, “SeedLegals’ customer service is so hands-on” when you have questions on the meaning of legal terms. While much of the legal work is done automatically, there is a step-by-step process of multiple-choice selection of terms to put to investors; the term selectable are standardised, based on market norms, but can be made custom on request if ever needed. There is always someone in the SeedLegals available on chat/email/phone to explain the process or the meaning of typical startup investment terms.
“It’s so helpful to work with SeedLegals, as you see the whole process mapped out in front of you, whereas when I was talking with law firms [as alternatives], there was a huge lack of clarity. Timings don’t get away from you so much when you have control of the process. So SeedLegals helps you to get your round closed on time too.”
“We’ve used a lot of the SeedLegals tools; from Founders Agreements, Consultant Agreements, and Employment Contracts, to NDAs. We like having all the documents in the one place, and that they are all synchronised. I love the digital cap table, it is very useful”.
“For EMI too, they’ve made it so much simpler.” This is a new solution for tech startups which dramatically simplifies the creation of EMI options for staff, with automatic document creation, and compliance checks. Much like Kortial’s own proprietary technology, SeedLegals EMI tool automates the complex legal and administrative work – empowering users to get more done, for less cost, in less time.
Startups like Kortical and SeedLegals are demonstrating the power of automation technology, heralding in a new era of Machine Learning SAAS that responds directly to customer needs. AI is becoming the new normal, delivering life-saving benefits in this instance, as well as massive productivity gains and data insights. It is exciting to watch. What radical new tools will Kortical give birth to in the next few years, as they expand their client base on the back of this recent investment? That really comes down to the imaginations of their clients.