Copenhagen, Capital Region of Denmark, Denmark
➕ rooted in mathematics 🤖 machine & deep learning enthusiast 💻 passionate pythonista 📈 free-time quant 🚀 forward-thinker 📎 github.com/VanekPetr With more than 5 years of experience in the field, I have firmly established myself as a versatile Machine Learning professional. Currently, I'm leveraging my expertise as a Machine Learning Engineer at Go Autonomous in Copenhagen, mainly focusing on Natural Language Processing (NLP) and Computer Vision. My role involves designing microservice architecture, developing core vision and NLP models, creating APIs, automating pipelines, and working on product development. Outside of work, I assist with teaching at DTU and KU, co-supervise theses related to OR, ML and asset allocation, and apply my knowledge to develop an open-source, algorithm-based financial advisor. I earned my MSc degree in Mathematics-Economics from the University of Copenhagen and completed two BSc programs in Mathematics and Economics & Finance at Charles University (Prague). During this time, I mainly focused my research on Operation Research theory, Machine Learning, and its applications in Finance - with a particular interest in Asset Allocation. I was fortunate to receive a full Chinese Government Scholarship, which allowed me to spend a semester at Tsinghua University in Beijing - China's top-ranked university. As an ardent Pythonist and Deep Learning enthusiast, I dream big and consistently stretch my skills to adopt new, cutting-edge technologies and adapt to ever-evolving industry trends. Having originated from a Mathematics background, I am committed to further deepening my Machine Learning expertise and bringing about innovation through my work.
Teaching Assistant for the intensive 4-week summer course of Practical Financial Optimization at University of Copenhagen. The course gives an introduction to the domain of practical financial risk and portfolio management as well as an introduction to the programming language GAMS (General Algebraic Modelling Systems). The main objectives of the course are: - Measuring and managing return and risk trade offs - Designing and solving financial optimization models - Immunization and dedication of a bond portfolio - Modelling Value at Risk and Conditional Value at Risk - Back-testing results of ex-ante optimization
At Go Autonomous, I lead the AI Implementation branch of our Machine Learning team, where we bridge cutting-edge AI technology with real-world customer needs. This role emphasizes close collaboration with the Customer Success team to understand customer-specific requirements, deliver tailored features, and fine-tune model performance to deliver exceptional results for each client. This position emerged as we expanded our capabilities through access to the AI supercomputer Gefion in Denmark and strategically reorganized our ML team to focus on both foundational model innovation and direct customer impact. Key responsibilities: - leading integration and implementation of AI solutions tailored to customer environments - collaborating with Customer Success to identify and address unique customer challenges - fine-tuning models to optimize performance for individual use cases - leveraging state-of-the-art computational resources for scalable AI solutions - fostering a culture of collaboration and technical excellence.
As the AI Product Owner at Go Autonomous, I have expanded my role while continuing as a Senior ML Engineer. This transition comes at a pivotal moment for our company, following the successful closure of a $10.3 million Series A funding round and our growth to a team of over 60 talented individuals. In this extended role, my primary contribution has been leading the implementation of Generative AI into our product suite, transforming how our solutions empower customers and drive business outcomes. I spearheaded the integration of cutting-edge generative AI models to elevate product capabilities, ensuring they align seamlessly with both customer needs and business objectives. Key responsibilities included: - strategically rolling out Generative AI: Leading initiatives to incorporate Generative AI, from ideation to execution, enhancing the intelligence and automation capabilities of our offerings - collaborating closely with cross-functional teams, including Customer Success and C-level executives, to align AI initiatives with strategic goals - defining and prioritizing features and requirements for AI products, maintaining a clear roadmap and timeline - leading user acceptance testing and feedback gathering to iterate and enhance AI solutions - contributing to long-term AI strategy and vision while participating in company-wide initiatives as part of the extended leadership team.
Following Go Autonomous's successful seed investment of 3.1 million euros, which has facilitated our expansion to over 30 full-time employees, I've been promoted to Senior ML Engineer. In this elevated position, I oversaw the advancement of our document understanding capabilities, vision models, and deployment strategies. During my tenure as Senior ML Engineer, I've: - spearheaded the evolution of our microservice architecture to ensure unparalleled scalability and efficiency┇Kubernetes, KServe, Knative, Triton, Docker, GCP - elevated our core vision and NLP models to new heights to deliver cutting-edge solutions┇transformer architecture, YOLOv8, BERT, spaCy, LiLT, PyTorch, TensorFlow - experimented with, fine-tuned, and deployed Large Language Models (LLMs), driving innovation in natural language understanding and generation┇Hugging Face, Mistral, LLaMA, FLAN-T5, Gemma - led the development and maintenance of mission-critical core APIs to uphold industry-leading standards of performance and reliability┇FastAPI, Prometheus, Grafana, Plotly, Sentry, - orchestrated the automation of data preprocessing and model training pipelines to streamline operations and boost productivity┇Airflow, Vertex AI, GitHub Actions, PostgreSQL, W&B, - fostered extensive collaboration across teams, seamlessly integrating ML solutions into our products and services to meet and exceed customer expectations, - driven the creation of highly impactful ML-based products that not only meet market needs but also generate substantial profitability, solidifying our position as a leader in the autonomous technology landscape.
Open-source Python framework for developing and backtesting asset allocation models. - Adopted by 200+ students at University of Copenhagen and DTU - Supported 7 theses in quantitative finance (graduate & undergraduate) - Presented at the 68th EWGCFM & Research in Options Conference, Khalifa University, Abu Dhabi (Dec 2023), in collaboration with MOSEK ApS
Quantitative asset allocation modeling which aims to develop a revolutionary platform for investing with ETFs. Research in financial optimization theory and application for optimal portfolios of assets. It requires: - financial modeling in Python and GAMS | Bootstrapping and Monte Carlo scenario generation - application of machine learning algorithms | clustering | minimum spanning tree method - managing historical ETFs data | due diligence screening | backtesting
The main intention of my research at AlgoStrata was to develop a framework to measure the performance of Danish mutual funds and design an algorithm based on mathematical modeling and machine learning methods which can replicate the best performing mutual funds in Denmark.
Teaching Assistant for two courses, Optimization in Finance and Financial Products, at Technical University of Denmark. The content of the courses is: - Introduction to financial markets and instruments - Financial modeling in GAMS - Classification of risks and their modeling (tracking, Value at Risk, Conditional Value at Risk) - Classical concepts (duration, convexity) and models for fixed-income and stock portfolios (immunization, dedication, Markowitz) - Interest rate and stock price models - Scenario-based modeling and stochastic programming