Dimiter Shalvardjiev

The company wanted to better support businesses by providing a platform to monitor how customers feel about their products and services and through actionable insights for the day-to-day operations using tailored dashboards and machine learning to increase customer retention and sales.

I've managed a team of up 15 elements undertaking this project from scratch. Added value:
• Reduction in the costs of operation and the total cost of ownership
• Reduction in number of requirements for the DevOps team
• Improvement in operational and data security
• Independent infrastructure supported by the cloud provider
• Usage on-demand
• Easier access to other cloud-hosted resources and databases, such as CRM and shared files
• Increased relevance of survey questions due to timely-delivery results in higher response rates

Based on the "momentum" investment philosophy, users could invest fiat and get a balanced spread of the top 10 best performing cryptos for the rolling month. The wallet was rebalanced every week or month, based on the customer's preference. Performance data was sourced from the three largest exchanges. The technical solution included .NET and Angular. Because of the small team size (five elements), I combined multiple roles over time, such as product manager, solution architect, and scrum master.

Marketing data is best sourced as freeform text, just as the consumer stated it at the survey point. This data is both noisy and unstructured, and text analytics are a necessary tool to make sense of it at scale. Specifically, sentiment analysis is a great way to sift through millions of feedback forms and generate insights and improvement suggestions for consumer products.

I led the development team in charge of this tool, acting as a product manager and software architect and providing vision and software architecture where applicable.

Co-led and co-architected the Customer Insights program, enabling on-demand analytics for the business on top of a brand new data warehouse. Supported business definitions of data objects, enabled forecasting capabilities, and spearheaded stakeholder management.

An electronic medical record (EMR) system was developed with privacy by design in mind. The user hosts and manages his own digital medial record on dedicated, fully encrypted hardware. Users can add medical conditions, treatment plans, test results and track chronic diseases like high blood pressure and type 1 diabetes. Patients can share a one-time URL with their doctor when going on a medical check-up, allowing better-informed decisions based on anamnesis and ad-hoc measurements and tests.