Two days of talks, all focused on JetBrains MPS. Missed it? Every talk is available on video.
The MPS Community Meetup brought together language engineers, toolsmiths, and practitioners building the next generation of domain-specific languages with JetBrains MPS. Over two days at Canon Production Printing in Venlo, speakers from JetBrains, F1RE, Canon Production Printing, Belastingdienst and specificlanguages.com shared what they are building and learning.
Nine talks across two days, all centred on JetBrains MPS: from AI and agents to LionWeb, build automation and running MPS in the browser.
Companies set up a display showcasing not just their MPS implementation, but the skills, workflows, and internal processes behind it, giving the community real insight into how organizations build and apply MPS day-to-day.
A big thank you to Canon Production Printing for hosting the MPS Community Meetup and for their wonderful hospitality. Their campus in Venlo made the two days possible.
Click a talk for the abstract and speaker bio, or watch the recording on YouTube.
Kocelot status: a DSL definition platform embeddable in your web applicationRadimir Sorokin, JetBrainsIn this talk, we will share the current status of Kocelot, a project developed at JetBrains. Kocelot aims to make it possible to embed DSLs directly into domain-specific applications. We believe that not only should the language be tailored to its domain experts, but so should the application in which it is presented. This is difficult to achieve with traditional language workbenches. Our goal is therefore to provide an SDK that enables developers to define a language and integrate it into their application stack with minimal effort. In the current milestone, we are delivering an SDK that supports applications built with Spring Boot, PostgreSQL, and Kotlin on the backend and React on the frontend. The architecture is designed with further integrations in mind. About the speakerRadimir works at JetBrains as the technical lead of Kocelot. He previously worked on JetBrains MPS and on bringing a DSL for biological knowledge to the web. | ▶ Watch video |
How to put your MPS project on the rails, and keep it thereSergej Koščejev, specificlanguages.comAutomation in an MPS project often begins with a build script copy-pasted from another project, with minimal adaptations. It compiles the project and runs some tests, but maintainers may struggle to understand the build and extend it. This talk presents a practical path from such an inherited build to maintainable project automation. Using examples from real MPS projects, we will establish a structured Gradle build and progressively add model checking, architecture enforcement, consistency and pending migration checks. We will also examine how MPS projects can make use of headless access to MPS models and how the checks can help maintain a project across several MPS versions simultaneously. You will learn how to
About the speakerSergej has been working with JetBrains MPS since 2015, specializing in the infrastructure of MPS projects. Over the years he has helped customers across a range of industries, including automotive, medical, and tax/payroll, with MPS performance, architecture, and migrations. He writes about MPS at specificlanguages.com and hosts free office hours for MPS developers at every level, from students to professionals. | ▶ Watch video |
Project dependencies for end users in MPSJohan Blok, BelastingdienstIn a sense, MDSE turns ordinary users into software developers. Inevitably, as the adoption of an MDSE solution grows, it encounters many of the same challenges as traditional software development, such as organizing a code base into multiple modules, some of which are more general in nature. Since our users are non-technical, conventional software engineering solutions cannot be applied directly. This presentation demonstrates a solution that enables non-technical end users to reuse their MPS projects without requiring them to specify build details. The solution extends the MPS build language with declarative constructs for specifying dependencies and the artifacts that are required or produced. By making a few assumptions, for example that each repository contains a single project, it becomes possible to generate build scripts for end-user specifications, including dependencies on specifications located in other projects, without requiring users to provide technical implementation details. The source code will be made available before the start of the conference. Demo outline: project dependencies for end users
About the speakerVarious software engineering jobs, among which experience in model driven software engineering using Xtext. PhD in history of logic and philosophy of mathematics. Senior model driven software engineer and product owner ALEF at Belastingdienst. | ▶ Watch video |
LionWeb in practice: a unified DSL across MPS, C# and TypeScriptNiko Stotz, F1REA manufacturer of industrial lasers needs to program its machines for diverse activities across a range of domains. Over time, several editors were developed for these activities, as the company nowadays provides not only devices, but integrated solutions. Due to the different focus areas, these editors use different representations of the underlying program, which complicates data exchange and close cooperation. As a solution, we developed a DSL capable of representing relevant aspects of a program in a unified language. We used MPS for prototyping, C# for backend development, and TypeScript for frontend implementation. LionWeb ties all technologies together. This talk presents the customer's challenges, the advantages and issues with DSL-based development, and how LionWeb underpins the solution. We present the language development process, and how LionWeb ideas helped in a real-world project: unified M2/M1 representation, generators and tools for C# and TypeScript, the bulk and delta protocol, language evolution, and derived models. About the speakerNiko heads the Model Driven Engineering team at F1RE and supports colleagues and customers with language engineering and DSL expertise. | ▶ Watch video |
Genie: agent support as a first-class aspect of MPS languagesAtanas Marchev, F1REAI systems are being connected to almost every modern coding platform. For general-purpose languages like C or Scala, this is a one-shot operation: every user will use the same language and functionality. In contrast, domain-specific languages, especially in JetBrains MPS, face the challenge of being highly user-specific, which makes useful and scalable AI integration a major challenge. In this talk, we present F1RE's agent-support DSL, Genie. Genie is a toolbox for defining and implementing agent-specific functionality in MPS. We implement the toolbox as an aspect, making AI support a first-class citizen of domain-specific language development. About the speakerAtanas is a Model Driven Engineer at F1RE, where he has worked since December 2021. | ▶ Watch video |
MPS and AI in Practice: Evidence from Real-World Application in Multi-Disciplinary Systems DevelopmentEugen Schindler, Canon Production PrintingThe rapid development of AI and agentic coding raises an obvious question for established modeling technologies: can they keep up? At Canon Production Printing, we have been putting some of the newest AI-enabling technologies around MPS to the test. We have piloted the mops CLI, PortaLon, and the Projectional Agent Toolkit on non-trivial use cases from our internal modeling activities as well as open-source language development. In this talk, we will show concrete examples of what we tried and what we learned. Our experience shows how MPS models and languages can be accessed and worked with beyond the traditional MPS IDE, how agentic coding technologies can become part of real modeling workflows, and how this creates new possibilities for integrating MPS with the wider engineering tool landscape at CPP. These technologies are still young and developing rapidly. But based on our practical experience with them, we believe there is already substantial evidence for a stronger conclusion: MPS has not merely kept up with the AI era. It is remarkably well positioned to take advantage of it. About the speakerEugen Schindler is Systems Architect Digital Transformation at Canon Production Printing and a long-time practitioner of domain-specific modeling and JetBrains MPS. He works at the intersection of multidisciplinary systems modeling, data, and AI, connecting models, engineering disciplines, computation, and people. He is part of the team building the Digital Twin Factory, spanning engineering models and DSLs, data, high-performance computing, AI inference, and agentic engineering workflows. | ▶ Watch video |
Release the mops: How to let coding agents loose on MPS projects without driving you crazySergej Koščejev, specificlanguages.comHave you ever tried asking a coding agent to implement a new feature in your MPS project, only to have it run in circles for half an hour and make no progress? An MCP server helps, but how do you set up everything so that an agent can do its work in a separate worktree in the background? What if there was a simple way to have an agent (or several) work on your MPS project in parallel with you or other agents without stepping on each other's toes? In this talk, I will demonstrate mops, a command-line (CLI) tool for MPS, and show how its architecture gives it capabilities beyond what most MCP servers can provide. You will learn how to
About the speakerSergej has been working with JetBrains MPS since 2015, specializing in the infrastructure of MPS projects. Over the years he has helped customers across a range of industries, including automotive, medical, and tax/payroll, with MPS performance, architecture, and migrations. He writes about MPS at specificlanguages.com and hosts free office hours for MPS developers at every level, from students to professionals. | ▶ Watch video |
Modeling for Everyone: Developing High-End Production Print Systems, 8 Years LaterEugen Schindler, Canon Production PrintingEight years ago, Canon Production Printing presented how we had established a solid foundation for an ecosystem of JetBrains MPS-based domain-specific languages within our software development organization. Since then, this approach has proven its value and expanded beyond software into the multidisciplinary development of our platforms and products. In this talk, we look back at how our MPS ecosystem has evolved over those eight years: how it grew, how it became embedded in our engineering landscape, and how we continue to derive sustainable value from it. We will also discuss the challenges that come with maintaining and evolving a substantial DSL ecosystem in a company developing complex, multidisciplinary products. What does it take to keep such an ecosystem healthy and valuable over the long term? Finally, we will share our perspective on what we need from MPS and its community going forward, and how we see our role in continuing this successful collaboration. About the speakerEugen Schindler is Systems Architect Digital Transformation at Canon Production Printing and a long-time practitioner of domain-specific modeling and JetBrains MPS. He works at the intersection of multidisciplinary systems modeling, data, and AI, connecting models, engineering disciplines, computation, and people. He is part of the team building the Digital Twin Factory, spanning engineering models and DSLs, data, high-performance computing, AI inference, and agentic engineering workflows. | ▶ Watch video |
A Path to the Web: Running MPS Languages in the Browser via JavaScript TranspilationNiek Hulsman, BelastingdienstOne of the well-known obstacles in language engineering is the growing demand for web-based interfaces. There is no straightforward transition path from an established project developed in a feature-rich language workbench such as MPS. In fact, the challenge is largely about distributing code between server and client. In the ultimate solution you want to declare functionality independent of how it is deployed. One common approach employs MPS as a backend, with only the user interface running on the client. For larger projects and user bases, however, this can result in quite significant server load. For a new product variant of Alef, we experimented with the opposite approach: running MPS on the client. This provides a feature-rich environment from the start and appears to be a better preparation for a transition path towards native web implementations, as well as a range of mixed server/client distribution architectures. We will demonstrate the Alef languages running in a browser, including editors, model checking, and a fully fledged interpreter and debugger. We use AI-assisted conversion of MPS source code and Alef languages to JavaScript using transpiler tooling. Since the technique is general and not Alef-specific, it may be of interest to other MPS users. We will sketch our approach, discuss the challenges encountered, how we addressed them, and outline how we aim to develop this into an automated and maintainable process. About the speakerNiek is a Model-Driven Software Engineer currently working on ALEF at the Dutch Tax Administration (Belastingdienst). Previously, he worked on model-driven development in the financial sector, using Xtext to design and implement domain-specific languages. | ▶ Watch video |