Stylianos Kampakis
Verified Expert in Engineering
Data Scientist and Machine Learning Developer
London, United Kingdom
Toptal member since January 17, 2019
Dr. Stylianos Kampakis is an expert data scientist living and working in London. He has a Ph.D. in machine learning from the University College London and an MSc in informatics from The University of Edinburgh. He is a member of the Royal Statistical Society and the UCL Blockchain Centre. His skills include machine learning, statistical analysis, deep learning, natural language, and Bayesian statistics.
Portfolio
Experience
- Machine Learning - 12 years
- Data Science - 10 years
- R - 10 years
- Python 3 - 8 years
- Statistics - 7 years
- Natural Language Processing (NLP) - 6 years
- Generative Pre-trained Transformers (GPT) - 6 years
- Deep Learning - 5 years
Preferred Environment
R, Python
The most amazing...
...thing is working on new interesting projects revolving around ML and AI.
Work Experience
Senior Data Scientist
Circadia
- Developed an ML-based algorithm to predict clinical outcomes.
- Wrote up and published a research paper for which I also did the analysis. The research paper concerned a medical study.
- Created a deep learning algorithm based on a radar device to detect bed exit activity.
Data Science Advisor
London Business School
- Advised students on various projects involving machine learning and AI.
- Taught students how to use R and Python to deliver data science projects.
- Worked alongside companies on scoping out projects for students.
Chief Data Scientist
Electi Consulting
- Created a recommender system using a deep neural network. Mentored a team of two data scientists on how to maintain the model and bring it to production.
- Built agent-based tokenomics models for token economies.
- Built a demand forecasting model for the Stoli group. Managed a data scientist and a software developer in order to bring this model to production.
- Created a classifier on text data using natural language processing.
- Created a machine learning classifier to classify fraudulent reviews.
Chief Data Officer
Nitrous
- Developed a complete automated data analysis system based on genetic programming and interpretable machine learning.
- Managed and mentored two junior data scientists who have tested and made improvements to the engine of the automated data analysis system.
- Ran educational workshops on data science for a non-technical audience, in order to communicate how data science can be applied in their business.
CEO | Chief Instructor
Tesseract Academy
- Taught executive education workshops for the US Navy, Vodafone, and British Land.
- Designed data strategy for multiple companies, mostly startups and scale-ups.
- Taught over 100 decision-makers on how to use data science in their business.
Visiting Lecturer
Cyprus International Institute of Management
- Taught statistics and quantitative methods to a group of over 100 students across multiple years.
- Taught social media analytics through the use of machine learning and AI.
- Taught AI and business strategy to multiple groups of executives.
Data Scientist
Brandtix
- Used natural language processing to extract insights from social media data.
- Used forecasting methodologies to predict an individual's growth of social media following.
- Ran anomaly detection algorithms on a time series of social media volume.
- Used topic modeling on social media data and created new algorithms for social media analytics.
- Built a statistical model that describes the worth of a football player, and connected this to a monetary value.
Machine Learning Engineer
Talent Worldwide Inc.
- Developed an AI customer chatbot that serves thousands of customers each month.
- Created an ontology that was used to enhance a product with AI capabilities.
- Built APIs that went into production, revised the specs, and improved them.
Senior Data Scientist
ASOS
- Developed a demand forecasting model using a blend of traditional and machine learning (ML) techniques.
- Built a funnel conversion model, which enabled the business to understand better how to tweak different parts of its pipeline.
- Advised on AI strategy, both on the use of large language models (LLMs) and small language models, as well as agentic orchestration.
Senior Data Scientist & Statistician
Osler Diagnostics
- Developed a variance decomposition model for a medical device.
- Assisted with the statistical design of an FDA submission.
- Designed and improved Streamlit dashboards for the analysis of biotech-related data.
Senior Data Scientist
Satalia
- Created a recommendation engine using a unique combination of LLMs and an Ontology framework.
- Created a system to automatically deliver marketing insights for a large retail brand.
- Designed and delivered an AI Agents framework for the extraction of strategic insights for the C-suite.
Senior Data Scientist
Osler Diagnostics
- Developed an analytics dashboard for experiment outcomes.
- Improved the statistical methodology for tracking and detecting outcomes.
- Created an AI strategy so the company can further enhance its capabilities.
Senior Data Scientist
Reskinned Clothing
- Designed and developed a dynamic pricing system that increased margins by 15%.
- Deployed and monitored a live machine learning system, and made improvements based on feedback.
- Managed a team of 2 people to deliver, deploy, and monitor the solution.
AI Advisor
Digipal AI
- Customized a large language model and trained it on GPUs.
- Created intellectual property for Digipal AI, which they use to raise funds.
- Created an AI and data strategy for the company, which is being used in fundraising.
Senior Data Scientist and AI Consultant
SquareTrade
- Used machine learning to create a customer churn prediction system.
- Created a time series forecasting algorithm for the logistics department.
- Helped create the AI and data strategy for the company.
Senior Data Scientist
Brainpool
- Delivered a forecasting algorithm to forecast sales for a London-based retailer and deployed it on Sagemaker.
- Set up a data standard for digesting and using data for predictive modeling.
- Developed a new machine learning method for forecasting. The method uses a combination of unique features.
Experience
ADAN.io
TokenLab
https://github.com/stelios12312312/TokenLabTokenLab is a comprehensive economic simulation framework built on agent-based modeling principles. Originally developed for digital economies, it has evolved into a versatile platform for analyzing any complex economic system where incentives, behaviors, and mechanisms interact dynamically.
Senior Data Scientist for WPP
1) Ontologies
2) Recommender systems
3) Predictive modeling
4) AI strategy
Education
Master of Science in Entrepreneurship in Business Administration
University of Cambridge - Cambridge, United Kingdom
Ph.D. in Computer Science
University College London - London, England
Bachelor of Science Degree in Statistics and Mathematics
The Open University - United Kingdom
Master of Science Degree in Informatics
The University of Edinburgh - Edinburgh, Scotland
Master of Science Degree in Intelligent Systems
Aristotle University of Thessaloniki - Thessaloniki, Greece
Diploma in Economics
The Open University - United Kingdom
Bachelor of Science Degree in Cognitive Science
Aristotle University of Thessaloniki - Thessaloniki, Greece
Skills
Libraries/APIs
XGBoost, API Development, Claude API, PyTorch, TensorFlow, CatBoost
Tools
Codex, ChatGPT, Claude, AI Prompts, Claude Code
Languages
Python, R, Python 3, SQL
Frameworks
LightGBM, LangGraph, Ontology Framework, Streamlit
Paradigms
ETL, Agile Software Development, Model Context Protocol (MCP), Management
Platforms
Harness, Databricks, Azure, Amazon Web Services (AWS), Docker, Kubernetes, Vertex AI
Storage
Data Pipelines, Google Cloud
Other
Machine Learning, Data Science, Statistics, Bayesian Statistics, Artificial Intelligence (AI), Deep Learning, Natural Language Processing (NLP), Forecasting, Statistical Forecasting, Education, Generative Pre-trained Transformers (GPT), Pricing Models, Supervised Learning, Classifier Development, Demand Forecasting, Excel 365, Data Analysis, Minimum Viable Product (MVP), Technical Leadership, Time Series, Fractional CTO, Recurrent Neural Networks (RNNs), AI Agents, Generative Artificial Intelligence (GenAI), Agentic AI, CTO, Early-stage Startups, Startups, New Business Ventures, Model Deployment, Linear Regression, Multi-agent Systems, APIs, Prompt Engineering, ChatGPT API, Predictive Modeling, Time Series Forecasting, AI Agent Orchestration, AI Consulting, Mentorship, Training, Team Mentoring, Reports, Reporting, Data Extraction, ChatGPT Prompts, AI Workflow, Technology, Training Workshops, Workshop Facilitation, Crypto, Agentic AI Systems, RAG Systems, Models, AI Evaluation, Analytics, Data-driven Marketing, Agentic RAG Systems, Causal Inference, Large Language Models (LLMs), Leadership, Conversational AI, Combinatorial Optimization, Strategy, AI Security, Business, Business Ideas, Finance, Supply Chain, Architecture, Supply Chain Management (SCM), Social Networks, Graph Neural Networks (GNNs), Ontologies, Clustering, Churn Analysis, Churn Management, Research, Simulations, AI Chatbots, AI Voice Agents, LangChain, ElevenLabs Solutions, Vector Databases, Retrieval-augmented Generation (RAG), Decision Support, Decision Support Systems (DSS), Hugging Face, OpenAI, Machine Learning Operations (MLOps), FastAPI, Light LLMs, GBM, EDA, Pricing, Survival Modelling, Dynamic Pricing
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring