# Projects — Jan Petrik > 18 projects across patent law, blockchain regulation and applied AI: patent prosecution and litigation, doctoral theses, and machine-learning research. Each links to its page on janpetrik.com. ## [Beyond parabolic weld bead models: AI-based 3D reconstruction of weld beads under transient conditions in wire-arc additive manufacturing](https://janpetrik.com/portfolio-AIBead.html) 2022 · Journal of Materials Processing Technology This study presents AIBead, an AI framework designed to predict and reconstruct a 3D geometry of weld beads in wire-arc additive manufacturing, especially under variable conditions where standard parabolic models are inadequate. It tackles the challenge of precise weld bead geometry prediction, offering an improvement by accurately depicting asymmetrical shapes along curved paths, thereby refining automated process planning and parameter optimization. - Type of Work: Scientific Publication - Main Author: Jan Petrik - Affiliation: ETH Zurich - Co-Authors: Benjamin Sydow, Markus Bambach - Date: 9th December 2021 - Journal: Journal of Materials Processing Technology - DOI: [10.1016/j.jmatprotec.2021.117457](https://doi.org/10.1016/j.jmatprotec.2021.117457) ## [Patent Prosecution](https://janpetrik.com/portfolio-patent_drafting.html) 2022–2026 · IPrime Rentsch Kaelin As a Patent Engineer and Patent Attorney Candidate at IPrime Rentsch Kälin AG, I drafted more than ten patent applications. I have since passed the European Qualifying Examination and am a qualified European Patent Attorney. These were mainly in the technical fields of IT, mechanical and electrical engineering. ## [Reinforcement learning and optimization based path planning for thin-walled structures in wire arc additive manufacturing](https://janpetrik.com/portfolio-RLPlanner.html) 2023 · Journal of Manufacturing Processes This research presents a path planning framework, RLPlanner, which uses reinforcement learning and Sequential Least Squares Programming optimisation for automatic deposition path planning in wire arc additive manufacturing of thin-walled structures. Addressing the limitations of existing path planning strategies, this paper demonstrates the framework's adaptability to geometry variations and its ability to adjust welding parameters. - Type of Work: Scientific Publication - Main Author: Jan Petrik - Affiliation: ETH Zurich - Co-Authors: Markus Bambach - Date: 5th May 2023 - Journal: Journal of Manufacturing Processes - DOI: [10.1016/j.jmapro.2023.03.013](https://doi.org/10.1016/j.jmapro.2023.03.013) ## [Learning Unsupervised Shape Recovery from Images](https://janpetrik.com/portfolio-masterthesis.html) 2019 · Master’s thesis, ETH Zurich The thesis presents an innovative approach to 3D shape reconstruction from images, exploiting the capabilities of a Generative Adversarial Network (GAN) called ShapeTexGAN. The research focuses on generating high quality 3D models from unordered point clouds or 2D images without requiring detailed camera information, with the aim of improving the efficiency and quality of 3D reconstruction processes where limited input data is available. - Type of Work: Master Thesis - Main Author: Jan Petrik - Affiliation: ETH Zurich - Supervisors: Radek Danecek, Markus Gross - Date: 20th October 2020 - Journal: None ## [Carbon Capture and Energy Storage](https://janpetrik.com/portfolio-bachelorthesis.html) 2017 · Bachelor’s thesis, CTU Prague Exploring the innovative intersection of technology and environmental science, this bachelor's thesis delves into the critical areas of Carbon Capture and Energy Storage (CCES). It addresses the escalating concern of atmospheric CO2 levels by presenting a comprehensive analysis of carbon capture methods and energy storage solutions. Through a blend of theoretical research and practical modelling, this thesis contributes to the ongoing dialogue on sustainable energy practices by developing a functional model capable of generating electricity from compressed air that simulates compressed CO2. - Type of Work: Bachelor Thesis - Main Author: Jan Petrik - Affiliation: CTU Prague - Supervisor: Vaclav Dostal - Date: May 2017 - Journal: None - Online: https://dspace.cvut.cz/handle/10467/69928?locale-attribute=en ## [Experimental Investigation of Cold-Spray Chromium Cladding](https://janpetrik.com/portfolio-cladding.html) 2018 · TopFuel This paper explores the effectiveness of cold-spray chromium (Cr) coatings on Zircaloy-4 for enhancing the accident tolerance of nuclear fuels. By focusing on optimized Cr coatings subjected to high-temperature steam tests, the research contributes valuable insights into accident-tolerant fuel (ATF) technologies. - Type of Work: Conference Publication - Main Author: Mohamed Shahin - Affiliation: MIT - Co-Authors: Jan Petrik, Arunkumar Seshadri, Bren Phillips, Koroush Shirvan - Date: 2018 - Conference: TopFuel 2018 - Online: https://www.euronuclear.org/archiv/topfuel2018/fullpapers/TopFuel2018-A0193-fullpaper.pdf ## [Fatigue Behavior of Cold Spray-Coated Accident Tolerant Cladding](https://janpetrik.com/portfolio-cladding2.html) 2018 · TopFuel proceedings This paper investigates the fatigue behaviour of cold spray coated Zircaloy cladding, a major advancement in Accident Tolerant Fuel (ATF) cladding for nuclear reactors. By examining the durability of coatings such as FeCrAl, Cr and Mo under normal and extreme conditions, the study highlights the importance of coating integrity in enhancing the safety and operational reliability of nuclear fuel cladding. - Type of Work: Conference Publication - Main Author: Martin Sevecek - Affiliation: MIT - Co-Authors: Mohamed Shahin, Jakub Krejci, Jan Petrik, Ronald G. Ballinger, Koroush Shirvan - Date: 2018 - Conference: TopFuel 2018 - Online: https://www.researchgate.net/profile/Martin-Sevecek/publication/328066444_FATIGUE_BEHAVIOR_OF_COLD_SPRAY-COATED_ACCIDENT_TOLERANT_CLADDING/links/5bb5f49aa6fdcc1b31321b0a/FATIGUE-BEHAVIOR-OF-COLD-SPRAY-COATED-ACCIDENT-TOLERANT-CLADDING.pdf ## [DeepForge: Leveraging AI for Microstructural Control in Metal Forming via Model Predictive Control](https://janpetrik.com/portfolio-deepforge.html) 2024 · Journal of Manufacturing Processes This study presents a method for microstructure control in metal forging that combines Model Predictive Control (MPC) with a newly developed machine learning model called DeepForge. Using 1D convolutional neural networks and gated recurrent units, DeepForge predicts microstructural changes from surface temperature measurements, providing a novel approach to improving forging quality and precision. By integrating real-time control via MPC, this research demonstrates the ability to dynamically adjust process parameters for optimised microstructural results. - Type of Work: Scientific Publication - Main Author: Jan Petrik - Affiliation: ETH Zurich - Co-Authors: Markus Bambach - Date: 25th February 2024 - Journal: Journal of Manufacturing Processes - DOI: [10.1016/j.jmapro.2024.05.023](https://doi.org/10.1016/j.jmapro.2024.05.023) ## [RLTube: Reinforcement Learning Based Deposition Path Planner for Thin-Walled Bent Tubes with Optionally Varying Diameter Manufactured by Wire-Arc Additive Manufacturing](https://janpetrik.com/portfolio-RLTube.html) 2024 · Manufacturing Letters This study presents RLTube, an algorithm that uses reinforcement learning (RL) to optimise deposition path planning in wire-arc additive manufacturing for the production of thin-walled bent tubes with variable diameters. Unlike traditional methods based on rigid mathematical rules, RLTube exploits greater flexibility, adaptability and efficiency by processing 2D images of tubes without the need for additional feature extraction steps. - Type of Work: Scientific Publication - Main Author: Jan Petrik - Affiliation: ETH Zurich - Co-Authors: Markus Bambach - Date: 3rd February 2024 - Journal: Manufacturing Letters - DOI: [10.1016/j.mfglet.2024.01.007](https://doi.org/10.1016/j.mfglet.2024.01.007) ## [Patent Litigation](https://janpetrik.com/portfolio-Litigation.html) Practice As a Patent Engineer and Patent Attorney Candidate at IPrime Rentsch Kälin AG, I have had the opportunity to participate so far more on two patent litigation cases in the Swiss Federal Patent Court. These were mainly in the technical fields of mechanical and electrical engineering. I have since passed the European Qualifying Examination (EQE) and am a qualified European Patent Attorney. ## [MeltPoolGAN: Auxiliary Classifier Generative Adversarial Network for melt pool classification and generation of laser power, scan speed and scan direction in Laser Powder Bed Fusion](https://janpetrik.com/portfolio-MeltPoolGAN.html) 2023 · Additive Manufacturing The study presents MeltPoolGAN, a machine learning architecture designed for the classification and controlled generation of melt pool images in additive manufacturing. It significantly improves the accuracy of classifying process parameters such as laser power, scan speed and scan direction, achieving around 97% accuracy for power and scan speed classification and less than 3 degrees of error for scan direction estimation. - Type of Work: Scientific Publication - Main Author: Jan Petrik - Affiliation: ETH Zurich - Co-Authors: Baris Kavas, Markus Bambach - Date: 10th November 2023 - Journal: Journal of Additive Manufacturing - Left: MeltPoolGAN generates melt pool images across 25 categories with a single rotation, covering a power range of 110 to 190 W and scan speeds from 200 to 1000 mm/s. - Right: Predictions of melt pools along the laser path, with scan speeds v1 at 900 mm/s and v2 at 600 mm/s. - DOI: [10.1016/j.addma.2023.103868](https://doi.org/10.1016/j.addma.2023.103868) ## [Development of Smart Experimental Facilities](https://janpetrik.com/portfolio-mit.html) 2018 · MIT During the seven-month stay at MIT, the focus was not on the development of new nuclear fuel cladings, where reliable and intelligent experimental facilities play a crucial role. The primary facility developed was aimed at measuring the burst pressure of pellets. A thermometric device for neutron imaging of two-phase flow was also developed. - Type of Work: Experimental - Main Author: Jan Petrik - Affiliation: MIT - Supervisors: Bren Phillips - Date: 2018 - Journal: None ## [CrystalMind: A surrogate model for predicting 3D models with recrystallization in open-die hot forging including an optimization framework](https://janpetrik.com/portfolio-CrystalMind.html) 2024 · Mechanics of Materials This paper presents CrystalMind, a surrogate model designed for fast and accurate 3D prediction of deformation and recrystallisation of workpieces undergoing open die hot forging in 3D. Using machine learning, specifically MLP (Multi-Layer Perceptron) and PointNET++ architectures, CrystalMind significantly reduces computation time compared to traditional approaches such as FEM simulations. - Type of Work: Scientific Publication - Main Author: Jan Petrik - Affiliation: ETH Zurich - Co-Authors: Syed Irtiza Ali, Martin Feistle, Markus Bambach - Date: 1st December 2023 - Journal: Mechanics of Materials - DOI: [10.1016/j.mechmat.2023.104875](https://doi.org/10.1016/j.mechmat.2023.104875) ## [Anti-Money Laundering Compliance without the Law: Can Blockchain Alone Do the Job?](https://janpetrik.com/portfolio-dissertation-edinburgh.html) 2024 · LL.M. dissertation, Edinburgh The LLM thesis explores whether blockchain technologies, through embedded economic incentives and social norms, can achieve regulatory equivalence with global anti-money laundering (AML) standards without relying on traditional legal enforcement. It critically examines the limitations of existing AML frameworks—particularly the FATF regime—and evaluates emerging alternatives enabled by distributed ledger technology. Drawing on Lawrence Lessig’s four modalities of regulation (law, architecture, market, and social norms), the thesis proposes a model of decentralized compliance by design. Through a simulated prototype and theoretical analysis, it assesses blockchain’s potential to align with core AML objectives while addressing legitimacy, accountability, and effectiveness concerns. - Type of Work: LLM Thesis - Main Author: Jan Petrik - Affiliation: University of Edinburgh - Supervisor: Dr. Morshed Mannan ## [Modelling of Manufacturing Processes via Machine Learning Algorithms](https://janpetrik.com/portfolio-dissertation.html) 2023 · PhD thesis, ETH Zurich This dissertation explores the use of machine learning techniques in manufacturing to enhance efficiency and quality. It investigates the real potential of AI through addressing three challenges: surrogate modelling, AI driven by experimental data, and reinforcement learning. Six AI models developed within this study tackle these issues. DeepForge and CrystalMind, as surrogate models, offer rapid prediction capabilities, notably reducing the time for optimizing forging processes from years to minutes. MeltPoolGAN and AIBead, empirical AI models, leverage real-world data to refine predictions in complex manufacturing tasks, with MeltPoolGAN notably improving melt pool image classification. RLPlanner and RLTube demonstrate the application of reinforcement learning in path planning for Wire Arc Additive Manufacturing, optimizing process parameters efficiently. - Type of Work: PhD Thesis - Main Author: Jan Petrik - Affiliation: ETH Zurich - Supervisor: Prof. Dr. Markus Bambach - Co-Examiner: Prof. Dr. Wolfram Volk - Date: 20th October 2023 - Journal: Six Publications in Journals - DOI: [10.3929/ethz-b-000662086](https://doi.org/10.3929/ethz-b-000662086) ## [Ubuntu Server](https://janpetrik.com/portfolio-server.html) Personal project This project describes the deployment and configuration of an Ubuntu Server 22.04 LTS operating system on a HPE ProLiant MicroServer Gen10 Plus V2. The primary motivation for this home server setup was to further my system administration skills, with a particular focus on environments that require high security standards for data management, such as patent or law firms. In these sectors, the integrity and confidentiality of data is paramount and the infrastructure must reflect the highest security measures. This server serves as a hands-on platform for exploring advanced system administration techniques, security protocols and data handling strategies tailored to the unique needs of legal and intellectual property environments. - Nextcloud: A self-hosted file share and communication platform. For patent and law offices, Nextcloud offers a secure environment for storing sensitive documents, collaborating on case files, and ensuring data privacy through end-to-end encryption. Its flexibility in access control and data sovereignty aligns with the stringent confidentiality requirements of legal practices. Jitsi: An open-source video conferencing solution. Jitsi provides a secure and private means for conducting virtual meetings, client consultations, and internal discussions without relying on third-party services. This is particularly relevant for maintaining attorney-client privilege and ensuring confidential communications in legal settings. GitLab: A web-based DevOps lifecycle tool that provides a Git repository manager. For patent and law offices, GitLab serves as an efficient platform for version control and collaboration on legal documents, development projects, and internal software, ensuring a traceable and organized workflow. Mail Server: A dedicated server for handling email communications. Implementing a private mail server ensures that all correspondence is managed securely, reducing the risk of sensitive information being intercepted or leaked. This is crucial for legal professionals who require confidentiality in their communications. RSS Feed: A tool for aggregating updates from preferred news sources and websites. For patent and law offices, an RSS Feed is invaluable for staying updated on the latest legal precedents, patent filings, and law-related news, aiding in research and keeping abreast of developments that could impact ongoing cases. Security Enhancements: The server is fortified with multiple layers of security measures to safeguard sensitive data. A robust firewall controls incoming and outgoing network traffic based on predetermined security rules, minimizing the risk of unauthorized access. Fail2ban monitors log files for signs of intrusion attempts and automatically enforces temporary or permanent bans on suspicious IP addresses. Two-factor authentication adds an extra layer of security, ensuring that only authorized users can gain access, while HTTPS encryption protects data in transit, making it difficult for intercepted data to be deciphered. Docker: Employed for containerizing applications, Docker enhances the server’s deployment and management efficiency. In patent and law offices, it ensures consistent environments for legal databases and client portals, minimizes downtime, and bolsters security through application isolation. This streamlined approach aids in maintaining high operational standards and data integrity. PreviousModelling of Manufacturing Processes via Machine Learning AlgorithmsAll projectsNextBlockchain Compliance and Regulatory Challenges in the Swiss Banks in a Global Setting Jan Petrik ## [Blockchain Compliance and Regulatory Challenges in the Swiss Banks in a Global Setting](https://janpetrik.com/portfolio-dissertation_legal.html) 2026 · Dr. iur. UZH, magna cum laude The PhD thesis explores the legal risks associated with the implementation of blockchain technology in the banking sector, with a particular focus on Swiss financial institutions. As blockchain continues to disrupt traditional banking infrastructures, it introduces a range of complex legal challenges—spanning regulatory compliance, data protection, cross-border transactions, smart contracts, and the classification of digital assets. This research aims to systematically analyze these risks in the context of both domestic and international legal frameworks. By investigating how existing laws apply to decentralized systems and identifying potential legal gaps, the thesis provides critical insights into the regulatory uncertainties banks face when adopting blockchain. Furthermore, it evaluates the role of global standard-setting bodies and national regulators in shaping the legal environment. Through this legal risk analysis, the thesis contributes to developing practical guidelines for banks seeking to integrate blockchain technologies while maintaining legal certainty, operational resilience, and regulatory compliance. - Type of Work: PhD Thesis - Main Author: Jan Petrik - Affiliation: UZH Zurich - Supervisor: Prof. Dr. iur. Rolf Weber - Co-Examiner: Prof. Dr. iur. Peter Nobel - Completion Date: -->