What I learned, and where I built it.

Majors in Data Science and Entrepreneurship.

Thesis "Bridging the Domain-Gap in Computer Vision Tasks": object detection and pose estimation for AR, trained on GAN-enhanced synthetic data. Graded 9.5/10.

Majors in Software Science, Entrepreneurship and Game Design.

Bachelor's thesis: development of a production-ready Meteor application for an external client. Final grade 9.

Subject profile Science & Engineering, supplemented with Biology and Computer Science.

Graduation project: an Arduino robotic arm, controllable from Android via a custom-built Bluetooth API. 9/10.

One project spanning the full breadth of the craft — computer vision, synthetic data, a geospatial front-end, VR/AR and DevOps. As a full-stack engineer on Project Gambit I help build AI that automatically detects anomalous behaviour in camera footage. Gambit is a building block of the Flexibel Reactie Concept (FRC), led by the Kenniscentrum Bewaken & Beveiligen (National Police, Public Prosecution Service, NCTV, Royal Netherlands Marechaussee).

• Built the back-end and front-end of the Gambit platform as a full-stack engineer, working closely with data scientists, designers and domain experts within a microservice architecture handling high-velocity live data and web streams. Delivered a high-performance front-end with geospatial data visualisation and interactive 2D drawing tools.
• Independently delivered the Algorithm Register end-to-end (back-end, front-end and DevOps) as an MVP to streamline AI Act compliance for defence algorithms and software, rolled out within Defence's internal software landscape and integrated with internal authentication providers.
• Built a synthetic data generator in Unreal Engine that produces automatically annotated training data for computer-vision models.
• VR/AR proof of concept in Unreal Engine: geospatial data integrated with live video streams.
• Helped shape the IT and AI roadmap for Project Gambit together with the Gambit team. Preparing demos, gathering requirements, and engaging with end users, stakeholders and software vendors.

Unreal Engine Python Vue 3 JavaScript TypeScript HTML SCSS OpenCV PostgreSQL RabbitMQ WebRTC WebSockets pytest SQLAlchemy Docker D3 Konva Cesium PrimeVue FastAPI Quasar Azure Pinia pgvector Genetec API Genetec Web SDK

MurrMurr lets organizations listen where that never used to scale. It runs personal, adaptive conversations with employees and customers and distils them into themes, sentiment and concrete next steps — the reach of a survey with the depth of a conversation.

• Designed and built the full platform: Vue 3 / Nuxt front-end, FastAPI (Python) back-end, Neo4j graph and PostgreSQL/pgvector — deployed on Azure with Terraform and Docker.
• Defined and built the AI roadmap: a multi-provider LLM agent system (OpenAI, Anthropic Claude, Google Gemini via LangChain) with RAG over vector embeddings and voice (TTS/STT).
• Multi-tenant SaaS with Auth0, org/role isolation and Stripe — type-safe via OpenAPI.
• CI/CD with GitHub Actions, tests (pytest, Vitest, Playwright) and AI-assisted (dev) workflows.

Neo4j GitHub Actions pytest Playwright Auth0 pgvector Terraform Azure Python TypeScript Sass PostgreSQL Vue 3 FastAPI MCP Servers LangChain LangGraph ElevenLabs API OpenAI API Claude API

Strategy is nothing without execution. At Pirlo I built the AI platform that finds the right growth strategy — and helps organizations actually act on it.

Neo4j Bitbucket Pipelines pytest Auth0 Weaviate Terraform Azure Python TypeScript shadcn-vue Sass PostgreSQL Vue 3 FastAPI LangChain LangGraph OpenAI API

Carried out the migration to AWS cloud for CI/CD processes, infrastructure and backend systems. Technically responsible for DevOps and advisory work on two TV applications (Geheugen Trainer (>50,000 active users) and Heel Holland Bakt).

Unity Xcode Android Studio AWS EC2 Unity Cloud .NET C# AWS Lambda

For EnergyWorx's energy-data platform I led the front-end migration from AngularJS to Vue — including architecture decisions, maintenance of the legacy codebase, new functionality, mentoring junior colleagues and hiring team members.

Vue JavaScript TypeScript HTML SCSS Vuex Cypress AngularJS Figma

Full-stack data scientist and software engineer at a machine-leasing startup. Developed several portfolio products in Vue, Laravel, GraphQL and React Native, including a backoffice platform and a big-data management platform with a custom front-end for annotation and data enrichment.

Specialised in computer-vision detection (YOLO, MobileNet, Xception), with an active role in shaping the AI vision and the roadmap for product integration.

Vue JavaScript TypeScript Laravel Redis Eloquent Python TensorFlow Keras PyTorch Unity HTML SCSS OpenCV NumPy GraphQL PostgreSQL Elasticsearch React Statamic C#

Research into state-of-the-art computer-vision algorithms, including YOLO and Convolutional Pose Machines, trained on GAN-enhanced synthetic data. Delivered an iOS prototype with on-device pose estimation for Philips consumer products in augmented reality.

Unity Python LaTeX Swift C Xcode

Design and implementation of an in-house software module, with ASP.NET as the back-end and Elm (a functional language compiling to JavaScript) as the front-end.

Elm ASP.NET C# HTML CSS JavaScript

Software developer at a collective of independent programmers and marketing specialists focused on scaling ambitious businesses.

jQuery JavaScript HTML CSS PHP Laravel Phaser

Built a .NET / Entity Framework application for file and data-resource management — now part of IB's core operations, reducing redundant data usage by 32%. Also built Symfony modules for ib.nl, including a WYSIWYG CMS with custom .less compilers and a parser that generates front-end templates from a self-designed declaration language.

HTML LESS CSS C# PHP Symfony .NET MySQL

Active in IT projects since the age of twelve — from simple websites to full database systems, frameworks and APIs. Brouns-design still serves as the umbrella for my client work.

HTML SCSS PHP Laravel PostgreSQL