Hi, I’m Sergio Verga – a Data Science Master's student (and external research collaborator) with a background in Physics (Microelectronics). I'm currently working on my thesis in Reinforcement Learning for Robotics at the University of Milano-Bicocca and doing research at the Intelligent Sensing Lab (ISLab). My main interests include: Reinforcement Learning (& Robotics), Deep Learning and Natural Language Processing. I'm passionate about bridging cutting-edge AI techniques with real-world applications, especially in automation, robotics, and intelligent systems. Let’s connect on LinkedIn or check out some of my work below!

• Vision-Language-Action (VLA) Models for Robotics
  Research on multimodal agents for robotic operations (e.g., pick-and-place of objects respecting particular features specified in the prompt). The tasks are studied using the Hugging Face LeRobot framework (part of my Master's thesis and research). [Currently private repositories]

• Semantic Segmentation on CamVid Dataset
  Deep learning pipeline for pixel-wise semantic segmentation using a NN from scratch (U-Net inspired), which has been optimized and compared to two models adapted via transfer learning (MobileNetV2 and DeepLab). GitHub Repo

• Reinforcement Learning Snake Game
  Simple RL agent trained to play the classic Snake game with Q-learning.
  GitHub Repo

• Automatic KPI Interpretation with KNIME + GenAI
  A KNIME solution for interpreting business KPI reports using workflows integrated with generative AI.
  View on KNIME Hub

• Automate KPI Report Interpretation and Insights with GenAI & KNIME
  Read the blog – Automates the interpretation of KPI dashboards with definitions, insights, and visualizations.

• 6 CEO KPIs and How to Measure Them
  Read the blog – A dashboard that tracks six essential business KPIs.

• Visualize Your Revenue Growth KPI with K-AI
  Read on Medium – A demo using KNIME AI Assistant to build quick custom visualizations.