logo

ai learn. ai create. ai impress

A data science team with emergent properties

Services

AI / ML solutions

Automation of services and processes as a business advantage. Various AI-tech approaches - ANNs, Evolutionary algorithms, Self-organization, Bayesian inference, and Expert systems.

Dynamic Optimization

Improving process performance and through dynamic optimization solutions. Using real-time data, the dynamic optimization process can help maximize predictive control. Helps by improving forecast results, and with AI included, forecasts can become even more accurate and predictable.

Customer Analytics

Customer analytics is focused on the examination of a company's customer information and customer behaviour. It aims to improve the identification of most profitable clients, as well as provides insights on the possibilities for attracting and retaining the most profitable customers. By using data-driven approach to customer analytics organizations can quickly establish successful customer interactions, and adjust its business practices in a most efficient way.

Behavioral Prediction

Predictive behavior modeling helps predict the future behavior of customers. By applying various techniques to historical and transactional data, prediction of future customer behavior is generated. Predictive behavior modeling goes beyond customer analytics by allowing companies to make decisions based on expected future results. Behavioral prediction helps stimulate cross-sell and improves the customer-brand relationship over time

Recommender systems

Providing personalized recommendations of products or services to users based on their previous preferences. Implementations through collaborative filtering methods, content-based methods, knowledge-based methods, ensemble-based methods, and evaluation.

Data warehouse

Enhancing business productivity by efficient and quick data gathering using cloud scalable solutions. Back-end tools development for ETL and data aggregation. ML model deployment in production as API.

Success stories

  • Propensity to buy

    Project “Next best offer”, DSK Bank

    Computing individual probabilities to buy a credit product, using statistical methods, to extract maximum sales potential from the data of own customers. Success: Increase of applications for sales by 99.8%

  • Auto Investment

    Research “Autonomous Investment Portfolio on Bulgarian Stock Exchange”, UNWE

    Analyzing the investment process as a cybernetic system where the portfolio (the controlled system) is given autonomy relying on self-organization. Success: 14-fold increase in risk-adjusted return

  • Customer segmentation

    Customer Segmentation Modelling with Machine Learning, A Logistics Company

    Creating customer segmentation model using unsupervised machine learning methods to help improve sales and marketing approach to different groups of clients. Success: Increased retention of customers

  • Quantum

    Quantum Algorithms

    Exploiting non-classical properties of the quantum computers to store and process exponentially large amount of information with applications in stochastic modeling. Success: Exponential speedup over classical computing.

  • Data engineering

    Data Extraction Service

    Real-time cryptocurrency exchange raw data extraction service, on top of which offline data warehouse was developed and integrated. Success: Quick, secure and reliable access for data teams to perform analysis, reporting and mining on the data.

  • Machine learning

    Machine Learning Competition, NumerAI

    It is a weekly data science competition for predictions, which steer Numerai's hedge fund strategy. Success: 348 NMR (~BGN 19.8k) earned

  • Time series

    Academia dataton

    Building a successful prediction model on the cryptocurrency markets for major cryptocurrencies prices in 48 hours. Success: 91.2% average accuracy (adj R-squared) for all cryptocurrencies, 2nd place among finalists.

  • Trading Algo

    Worldquant

    Worldquant challenge is where we have developed alphas - high performance algorithms for stocks trading. There were about 10,000 teams from all over the world, 11 teams were admitted to the second stage. Success: IQC Stage 2 virtual event


Team

Person's Photo

Milenka

Milena is a PhD student in Modeling and Forecasting. She is a financial analyst with 3+ years of experience. Experienced in interpreting and analyzing data to drive growth for a leading financial data provider. She has strong ability to identify, analyze and solve all kinds of problems, passionate about improving things and finding the best solution for everyone.

Person's Photo

Alex

Alex is a PhD student in Economics. She currently works as a project coordinator in the energy sector. Having experience with different organizational structures and business challenges, she is interested in applying data-driven solutions for resolving such challenges, as well as developing new business models using a data-oriented approach.

Person's Photo

Stefi

Stefi is a forex trader who is actively engaged in trading on the Forex market. Develops strategies for trading with currencies. Develops expert systems for trading with cryptocurrencies.

Angel Marchev

Angel

Prof. Angel Marchev, Jr. considers himself a fusion of thought explorer and a digital native. He is also a coeval and very good friend with the PC since 1980, when he started coding at the age of 4. More recently Angel earned a PhD in computer simulation and modeling for business. He has specialized at European mathematical society, while he is also a certified SME consultant by the Instiituut voor het Midden en Kleinbdrijf.

Pepe

Pepe

Pepe is a PhD student in Application of Quantum Computers, writing quantum algorithms with applications in different fields of cybernetics. Along dealing with quantum physics phenomenons like superposition and entanglement, he is developing software for market forecasting and designing technical solutions for various data science projects.

Person's Photo

Vasko

Vasil is PhD student in Modeling and forecasting. Hi is financial expert with 9+ years of experience. He is developing financial models about forecasting and reporting. His interests are related with implementation of data-driven solutions in financial management.

Person's Photo

Bogo

Bogo is a passionate programer with interests in many fields. He has 5+ years of experience as a data collection expert in various marketing research companies. Has experience with neural network algorithms for prediction modelling, image recognition algorithms, text mining, data scraping and process automation

Person's Photo

Sergo

Sergo is a natural problem solver who always seeks for an optimal solution. He is fascinated how different businesses operate, loves finding new insights and is passionate about data science. Together with his solid business background Machine Learning and AI algorithms become a real weapon in his hands for resolving various business data challenges.

Person's Photo

Marin

Marin is a PhD student in Modeling and Forecasting. CEO at MMQuants. His moto is: "Work smart and trade for a living, not live for trading"

Contact & Follow