Roel van Bergen
Zen Software Engineer | Team Rockstars IT
Gouda, The Netherlands
Actions
I’m Roel van Bergen: a software engineer, coach, speaker, zen teacher and proud father who thrives on both creating and connecting. When I’m not mentoring or crafting solutions in code, you might find me immersed in epic (board) games, cycling through scenic trails, or assembling intricate Lego sets. Fun fact: I love teaching, but my favorite lessons are the ones where I learn just as much from others.
I draw inspiration from my journey of continuous learning and reflection. With a background rooted in a zen buddhist approach to life, I aim to stay grounded, embrace growth, and celebrate the process as much as the results.
Area of Expertise
Topics
The Agentic Engineer: Making Implicit Engineering Explicit
How the Engineering Triad helps developers move beyond vibe coding and build environments where AI agents can safely contribute
---
AI agents can write code, run tools and move fast through a codebase. But speed alone does not create reliable software. In real development, generated code is not delivered work. A change only becomes valuable when it survives validation, review, acceptance and delivery.
This talk starts with a simple observation: onboarding a developer into a serious software project takes time. They need domain knowledge, architecture, test strategy, team conventions, delivery flow and context. So why do we expect an AI agent to understand the same environment from one vague prompt?
The real problem is not just that models hallucinate. It is that much of our engineering is implicit. Intent lives in conversations. Rules live in people’s heads. Validation lives in habits. Context is scattered across docs, tickets, APIs, tests and architecture decisions. Agents expose those weak spots quickly.
In this session, we explore the Engineering Triad as a practical mental model for agentic development:
Prompt Engineering makes intent explicit.
Context Engineering makes knowledge explicit.
Harness Engineering makes execution explicit.
Using a concrete password reset example, we will show how a vague request like “add password reset” hides security rules, API contracts, token handling, email infrastructure, validation strategy and human acceptance decisions. We will turn that into explicit artifacts such as a SPEC.md, BDD-style scenarios, AGENTS.md rules, scoped custom agents, reusable skills and validation gates.
The key message: agentic development is not about replacing engineering discipline. It makes engineering discipline more important. A good environment for agents is also a good environment for developers.
Target audience
This talk is for software developers, tech leads and engineering-minded architects who have already experimented with tools like ChatGPT, Copilot, Cursor, Claude Code, Codex or similar coding agents.
This is not an introduction to LLMs. It is for people who have felt the magic of AI-assisted coding and now want to make it reliable enough for real software delivery.
Working like a buddha: less stress more joy
Burnout is rising in tech, and many developers find themselves context-switching all day coding, Slacking, deploying, meeting, repeat. This interactive talk offers a down-to-earth introduction to mindfulness, grounded in Buddhist principles like attention. With relatable stories and practical tips, you’ll learn how doing one thing at a time can reduce stress, improve focus, and help you rediscover flow in your work
Orchestrating a game dev team using Microsoft Agent Framework
What happens when a software engineer wants to build a game but doesn't have a studio? They build the studio themselves: out of AI agents.
In this session, I’ll take you under the hood of a fully functioning game development team powered by the Microsoft Agent Framework. We will move beyond simple "hello world" prompts and explore the complexities of multi-agent orchestration. You’ll learn how to define specialised roles, manage inter-agent communication, and maintain a flow in decentralised workflows.
What you will learn:
Architecting the Team: How to assign roles (Designer, Coder, QA) within the framework.
Scaling Without Chaos: Lessons learned and "gotchas" when integrating multiple agents at scale.
The Reality Check: Practical insights into where agents shine and where the human touch remains irreplaceable.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top