Toon Weyens, Ph.D.

Utrecht, NL  ·  weyenst@gmail.com

Bridging science, data, and business to build AI solutions that drive real impact.
AI Solutions Architect with more than a decade of experience across pre-sales, enterprise AI platforms, and scientific High-Performance Computing (HPC). Strong expertise in building relationships of trust and navigating complex technical discussions. Experience in architecting data and AI solutions tailored to client objectives, while providing critical feedback directly from the field to R&D and product teams. Deep technical proficiency in HPC, distributed and cloud computing, as well as modern ML/DL techniques, coding and scripting methods. Passionate about building the solutions of the future by transforming customer needs into innovative, scalable, cost-efficient AI implementations that drive real business value and measurable impact. Committed to fostering lasting technical relationships that ensure sustained success and long-term growth throughout the customer journey.

About

Values:
Meaningful work · Authentic relationships · Radical transparency · Curiosity and experimentation · Ownership and excellence · Efficiency · Positive impact
Abilities:
Empathic and communicative · Analytical · Culturally adaptive · Scientifically rigorous · Self-motivated and Proactive · Collaborative
Nationality:
Belgian

Experience

Mar '24 - Current
Senior Sales Engineer - Anomalo
Partner with C-level and senior data leaders at global enterprises and AI-native technology companies to ensure trust in their data at scale. Combine data science, architecture, and business consulting expertise to help organizations automate data quality, strengthen governance, and accelerate AI adoption.
  • Led multi-stakeholder PoCs and enterprise evaluations with customers including BP, Kingfisher, Zalando, and others — representing over €4.5M in multi-year enterprise contracts.
  • Engage directly with executive sponsors (CDOs, CIOs, Heads of Data) to define success metrics and design scalable, compliant data-trust architectures.
  • Navigate enterprise sales cycles involving procurement, security, and data governance teams to accelerate platform adoption.
  • Champion platform capabilities across technical and business audiences, including applications of Generative AI for data quality automation and explainability alongside classical machine learning approaches.
  • Collaborate cross-functionally with product, R&D, and customer success teams to translate field insight into roadmap impact. Special focus on Anomalo's new Unstructured product which leverages genAI to create agentic DQ.
Jan '25 - Current
University Lecturer - Eindhoven University of Technology
Teach in both the Bachelor in Applied Physics and the Master in Nuclear Fusion programs, covering plasma physics and computational methods for High-Performance Computing (HPC).
Nov '22 - Mar '24
Presales Solutions Architect > Global Team Lead - Dataloop
Enabled large enterprises and AI-native organizations to operationalize data pipelines for computer vision, NLP, and multimodal AI at scale. Combined deep technical expertise with business acumen to lead complex, high-value engagements and mentor a global presales team.
  • Generated approximately €1M in new annual recurring revenue (ARR) over 18 months by leading enterprise AI data initiatives across automotive, defense, insurance, and telco sectors.
  • Led the global Presales Solutions Architecture team, aligning regional priorities, sharing best practices, and elevating technical sales performance company-wide.
  • Partnered with senior technical and business leaders to define architecture, KPIs, and success criteria for strategic AI programs.
  • Designed methodology for full-cycle PoCs and solution evaluations, ensuring alignment between business objectives and technical feasibility.
  • Collaborated closely with R&D and Product Management to shape platform evolution based on customer feedback from the field.
  • Created technical content and enablement materials that improved sales efficiency and shortened enterprise sales cycles.
  • Built and maintained strategic relationships across both technical and executive levels to secure long-term customer success.
Oct '21 - Nov '22
Customer Facing Data Scientist, Pre-Sales - DataRobot
Joined DataRobot's European pre-sales organization during its global expansion phase, working alongside some of the industry's most experienced enterprise AI professionals. Helped large organizations accelerate their AI maturity by translating complex machine learning capabilities into clear business value.
  • Led technical PoVs for enterprise customers in utilities, insurance, and manufacturing, connecting AI initiatives to strategic KPIs and ROI.
  • Collaborated with account teams and global sales leadership, learning and applying best-in-class enterprise sales methodologies (MEDDICC, consultative storytelling, value framing).
  • Partnered with IT directors and data science leaders to align AI strategy with existing data infrastructure and governance frameworks.
  • Delivered executive-level demos and workshops that demonstrated how to scale AI safely and effectively across the enterprise.
  • Provided structured field feedback to R&D and product teams to inform roadmap priorities and improve enterprise readiness.
Jan '19 - Aug '21
Industrial Data Science Consultant - MathWorks
Played a key role in expanding MathWorks' footprint from traditional individual and departmental licensing toward enterprise-scale data science and engineering platforms in the Benelux region. Partnered with both R&D engineers and IT leaders to enable scalable, secure, and collaborative environments for MATLAB and Simulink users.
  • Drove over €1.5M in additional annual recurring revenue (ARR) by establishing server-based enterprise data science platforms across key industrial accounts.
  • Led strategic customer engagements to design and implement modern, browser-accessible data and engineering environments integrated with live data streams and CI/CD pipelines.
  • Advised internal and external stakeholders on architecture, deployment, and best practices to ensure scalability and compliance.
  • Enabled cross-functional collaboration by connecting internal experts, mentoring colleagues on emerging technologies, and aligning resources for strategic accounts.
  • Championed MathWorks products through public speaking, webinars, and workshops, including the Deep Learning with MATLAB series.
  • Specialized in Data Science, IoT, Parallel and Cloud Computing (AWS & Kubernetes certified), High-Performance Computing, Enterprise Integration, and Computational Physics.
Jan '17 - Dec '18
Postdoctoral Monaco Fellow - ITER Organization
Conducted advanced research in plasma physics and magnetohydrodynamic stability at the world's leading nuclear fusion project. Awarded the prestigious Monaco Fellowship.
  • Published multiple first-author papers in peer-reviewed journals advancing understanding of plasma edge stability.
  • Investigated 3D effects on Edge-Localized Mode (ELM) stability, combining analytical modeling and HPC simulation.
  • Applied and extended the numerical code PB3D to model nonlinear plasma behavior in fusion devices.
  • Collaborated with international research teams to validate simulation outcomes against experimental results and support ITER design objectives.

Education

2012 - 2016
Ph.D. in Plasma Physics - Universidad Carlos III de Madrid · TU/e · ITER
Advanced research on plasma stability and 3-D effects in magnetic confinement fusion.
  • Authored multiple first-author publications in quality peer-reviewed journals.
  • Designed research project to improve understanding of high-n instabilities important for toroidal magnetic confinement devices for nuclear fusion.
  • Developed dedicated mathematical theory (Weyens et al, 2014, P.o.P, 21, 4).
  • Designed optimized numerical code, PB3D (Weyens et al, 2017, J.c.P, 330).
  • Became expert in modern Fortran and high-performance parallel computing techniques.
2010 - 2012
Master of Science - Nuclear Fusion Science & Technology - Ghent University · UC3M · Université de Lorraine
European Erasmus Mundus joint master's program in fusion engineering and science.
  • Studied advanced plasma physics and reactor design with interdisciplinary coursework across three European universities.
  • Engaged in international academic collaboration with strong focus on cultural and linguistic immersion.
  • Graduated in top 5% of class.
2008 - 2010
Master of Science - Energy Engineering - University of Leuven · TU Berlin
Multidisciplinary engineering program focused on energy systems, economics, and sustainability.
  • Studied integrated thermomechanical, electrical, and economic analysis of energy systems.
  • Completed academic exchange at TU Berlin, focusing on applied energy economics.
  • Graduated in top 15% of class.

Professional Development

2025
Autumn HPC School 2025 - Supercomputing center at the TU/e
Focused mainly on deep dive into GPU computing, learning how to accelerate Python workloads. Starting from an understanding of the fundamentals and comparison of the performance of scientific applications to optimization of parallel execution and identification of performance bottlenecks with profiling techniques. Special focus on deep-learning workloads with Large Language Models (website).
2025
Fine-tuning & RL for LLMs - AMD (DeepLearning.AI)
Studied post-training techniques for LLMs under Sharon Zhou. Covered alignment via Supervised Fine-Tuning and RLHF, advanced optimization algorithms like PPO and GRPO, and evaluation frameworks to detect reward hacking. Implemented efficiency methods such as Low-Rank Adaptation (LoRA) and built production pipelines for continuous feedback (cert).
2025
Agentic AI - Andrew Ng (DeepLearning.AI)
Focused on building agentic AI systems including tool use, planning, and multi-agent coordination. Implemented robust evaluation frameworks and reflection patterns for iterative improvement. Designed systems connecting to external APIs and databases, utilizing planning algorithms to execute complex workflows across specialized agents (cert).
2024
Mastering LLMs for Developers - Dan Becker, Hamel Husain (Maven)
State-of-the-art course on LLM development. Covered fine-tuning open-source models with Axolotl, instrumentation using Inspect and LangSmith, and iterative RAG design. Applied scaling techniques like PyTorch FSDP and torchao for efficiency, and deployed prototypes using Gradio, Modal, and Hugging Face AutoTrain (cert).
2018
Computability, Complexity & Algorithms - Georgia Institute of Technology (Udacity)
Computational theory including languages, countability, and Turing machines, plus implementation of advanced algorithms such as dynamic programming, FFT, and maximum flow (website).
2018
Bayesian Methods for Machine Learning - Higher School of Economics Moscow (Coursera)
Applied Expectation-Maximization, Variational Inference, and MCMC for probabilistic modeling; implemented Variational Autoencoders and Gaussian Processes using PyMC3, GPy, and GPyOpt; graduated with honors (certificate).
2018
Deep Learning - Andrew Ng (Coursera)
Studied deep learning foundations, convolutional and recurrent neural networks. Implemented CNNs, RNNs (BiLSTM, GRU), and sequence models in TensorFlow and Keras. Applied optimization and regularization techniques including Adam, AdaMax, BatchNorm, and dropout (certificate).
2014
23rd Summer School on Parallel Computing - CINECA
Intensive graduate program on modern high-performance computing systems covering parallel architectures, MPI, OpenMP, profiling, and debugging, with hands-on code optimization (website).

Projects

2013 - Current
PB3D - Peeling-Ballooning in 3-D
Developed a modern high-performance Fortran code for analyzing peeling-ballooning stability in toroidal magnetic confinement devices. Designed for high-performance and parallel computation supporting general 3-D plasma configurations; applied in nuclear fusion research for studying high-n instabilities and plasma edge stability. Website: PB3D.github.io. For subject matter experts:
  • Mathematical minimization of functional, leading to a generalized eigenvalue equation.
  • High-n instabilities excited by extreme temperature and pressure gradients from plasma to reactor walls.
  • General 3-D configurations and perturbed plasma edge, including resonance effects.
  • Postdoctoral research investigated 3-D effects such as RMP for ELM control or toroidal field coil ripple.
2018
Pylgrim
Implemented Python and C++ algorithms for the NP-hard Elementary Shortest Path Problem (ESPP). Benchmarked and improved upon recent academic methods from Di Puglia Pugliese (2016) and Boland (2006). Explored computational complexity and optimization strategies for constrained routing. Source: GitHub.
2018
Kraemer
Co-created an automated trading framework using Python for high-frequency arbitrage across exchanges. Integrated mathematical modeling, deep learning, and financial data engineering with a focus on efficient order execution and latency-sensitive decision-making.
2018
Facial Composits: Finding the Suspect
Capstone project for Bayesian Methods for Machine Learning. Co-created an automated trading framework using Python for high-frequency arbitrage across exchanges, integrating mathematical modeling, deep learning, and financial data engineering with a focus on efficient order execution and latency-sensitive decision-making.

Awards & Honors

2021
MVP Award - DataRobot
Most valuable award for Pre-Sales (DataRobot).
2019
Monaco/ITER Postdoctoral Fellowship - Principality of Monaco
The Monaco/ITER Postdoctoral Fellowship Program enables young researchers to contribute to fusion energy science within the ITER framework, working closely with leading experts in a unique international setting. Website: ITER.
2017
Ph.D. Research Award - European Physical Society
Up to four prizes annually recognizing truly outstanding Ph.D.-level research achievements in plasma physics across 38 European countries. Website: EPS.
2012-2016
Doctoral Scholarship - Universidad Carlos III de Madrid, CINECA
PIF scholarship for four years, including summer school funding.
© 2025 Toon Weyens   •  Theme adapted from  Moonwalk