19.10.2026 aktualisiert
100 % verfügbar
Machine Learning Engineer | Reinforcement Learning & Optimization Expert
Tirana, Albanien
Nur Remote
Master of Science in Machine Learning and Optimization, University of Technology of CompiegneSkills
Computer VisionPythonLineare OptimierungMachine LearningMathematische OptimierungPyTorchDeep LearningOptimization AlgorithmsDynamic Programming
Machine Learning & Deep Learning
Expertise in classical Machine Learning algorithms such as Random Forests, Support Vector Machines and Bayesian statistics, as well as Deep Learning techniques including Computer Vision tasks like classification, segmentation, localization and Variational Autoencoders using the PyTorch framework.
Reinforcement Learning & Optimization
Strong background in Reinforcement Learning with policy methods such as PPO, combined with Operations Research and Optimization techniques applied to real-world problems like offshore supply vessel planning and VNFs reconfiguration.
Data Science & Time Series Analysis
Proficiency in Data Science, probability and statistics, Time Series Forecasting and Classification, including Dynamic Time Warping and Conformal Prediction frameworks, applied in domains such as renewable energy and match prediction.
Python Programming
Advanced Python programming skills including object-oriented programming, use of simulation libraries such as SimPy, and data visualization with Plotly Dash.
Computer Vision
Application of Computer Vision models for object detection, segmentation, classification and localization, including the use of Variational Autoencoders.
Signal Processing
Knowledge and application of signal processing techniques in software engineering contexts.
C# & .NET Framework
Experience with C# programming and the .NET framework, including SQL and Design Patterns for software development.
Dynamic Programming & Combinatorial Optimization
Application of Dynamic Programming with approximation frameworks to tackle NP-Hard Combinatorial Optimization problems, including dynamic Beam Width and Expansion factor strategies.
Linear Programming & Scheduling
Advanced techniques in Linear Programming, Optimization, Scheduling and Complexity, as well as Uncertainty and Belief Functions acquired during academic training.
Expertise in classical Machine Learning algorithms such as Random Forests, Support Vector Machines and Bayesian statistics, as well as Deep Learning techniques including Computer Vision tasks like classification, segmentation, localization and Variational Autoencoders using the PyTorch framework.
Reinforcement Learning & Optimization
Strong background in Reinforcement Learning with policy methods such as PPO, combined with Operations Research and Optimization techniques applied to real-world problems like offshore supply vessel planning and VNFs reconfiguration.
Data Science & Time Series Analysis
Proficiency in Data Science, probability and statistics, Time Series Forecasting and Classification, including Dynamic Time Warping and Conformal Prediction frameworks, applied in domains such as renewable energy and match prediction.
Python Programming
Advanced Python programming skills including object-oriented programming, use of simulation libraries such as SimPy, and data visualization with Plotly Dash.
Computer Vision
Application of Computer Vision models for object detection, segmentation, classification and localization, including the use of Variational Autoencoders.
Signal Processing
Knowledge and application of signal processing techniques in software engineering contexts.
C# & .NET Framework
Experience with C# programming and the .NET framework, including SQL and Design Patterns for software development.
Dynamic Programming & Combinatorial Optimization
Application of Dynamic Programming with approximation frameworks to tackle NP-Hard Combinatorial Optimization problems, including dynamic Beam Width and Expansion factor strategies.
Linear Programming & Scheduling
Advanced techniques in Linear Programming, Optimization, Scheduling and Complexity, as well as Uncertainty and Belief Functions acquired during academic training.
Sprachen
EnglishverhandlungssicherFrenchGrundkenntnisse
Projekthistorie
Time Series Forecasting and Classification, Data Analysis in the Renewable Energy Domain, Data Science and Statistics.
Tackling the VNFs reconfiguration problem as an NP-Hard Combinatorial Optimization problem using Dynamic Programming with approximation framework. Dynamic Beam Width and Expansion factor applied; results compared against autoregressive Encoder-Decoder architecture trained with Reinforcement Learning.
Exploiting a large dataset of match coefficients series until match start time, Dynamic Time Warping for Time Series Classification predicts match outcomes. Conformal Prediction framework is employed for selective classification.