My application experience covers Financial (Asset , Wealth Management and Banking sector), Managed Healthcare, Pharmaceutical, Mining, Retail, Broadcasting, HR, Publishing and Transport Sectors where I have fulfilled the roles of Developer, Team Lead, Lead Architect, CIO , IT Manager and Data Scientist. I have lead teams from as big as 100 people and as small as 2. My project/development methodology I prescribe to are Agile and XP. I am a big proponent of DEVOPS , CD/CI pipelines for BI , machine leaning and software projects as this is often overlooked but should always be used in today’s competitive market and really a big component to get to market quickly. For the last 20 years I have been extensively using the Microsoft full stack for designing, developing and implementing on Premise and Cloud based Business Intelligence solutions. As much as I have specialized in the Microsoft stack of products I also have good knowledge and project experience of other BI tool-sets which give me a great ability to choose the correct architecture for each client and their business intelligence requirements, also using what the client already has. I have skilled myself to use statistical modelling and to also fulfill a data scientist role to benefit the customer by using data driven decisions and closing the loop by integrating these models into line of business systems to make decisions and recommendations. Early successes – 2000 to 2010 1. I architected the first commercially available real time medical claims processing system (biggest) in the world at the time – by removing the manual processing of 500,000 documents a month the company saved approximately 5 million rand a month. I traveled to various countries to showcase the solution at various conferences. 2. I have designed and developed the only bespoke asset management Forecasting tool for an Asset and Wealth management business using the native Microsoft stack of products and have been invited by Microsoft as a speaker to showcase the solution. The cost savings for this application was calculated at 4 million rand saving per year using fully automated processes maintained by one user compared to 400 users doing the work manually. The time to forecast took 3 days per person per month per portfolio or fund and another 5 days to collate all the data before a forecast was complete and view-able. This was done multiple times manually to get a tweaked forecast until CFO was satisfied. The new solution was able to forecast revenue for all their portfolios and funds within minutes and could be run multiple times with new scenarios - the tool was metadata driven and are able to stay relevant and adapt to new requirements. The system is still in use and been running for 8 years now. Latest notable Successes 2011- now 1. Developed an Artificial intelligence chat bot that allowed pediatricians to build scenarios that allows the AI model to continuously adapt. The back-end systems are cloud based and is accessed via mobile platforms for parents to use. It has various integration endpoints and allows the building of a medical profile for a personalized experience for each patient. The solution is due to go live in next couple of months. This will rival products like WEBMD but targeted at Parents. This is built for the South African market as additional services will be based on locally available products and medical practitioners. This app also allows live/chat access to pediatricians or other medical practitioners. 2. Developed a Dental care management system with COLGATE to provide a patient a better way to care for dental needs as most medical aids have a limit on expenditure for dental and have no need to manage it for the member. The system was driven by clinical protocols and was proven to save money not only to the member but also savings for the medical aid as better dental care prevents additional disease as oral health is a window to your overall health. This motivated Momentum Health to purchase the complete framework. 3. The problem a car rental company had was that their fleet of cars was not being utilized effectively - they could only achieve a 68-72% utilization. They used spreadsheets and made certain assumptions and growth calculations to move the fleet to a branch for rental. Effectively some branches did not have enough vehicles to rent and some branches had vehicles not needed and effectively lost income and incurred cost for the storage of vehicles not used. Using a Machine learning model algorithm that used multi seasonality with 5 years of rental data we were able to effectively determine location and demand. The result of using the AI model increased the fleet utilization up by 12 %. We had an added benefit as we were able to determine the size of their fleet and when demand required vehicles was leased for a limited time. They were able to sell 3000 vehicles and save the costs related to owning and maintaining 3000 vehicles.