VIVENTRIS – Smart production solutions for the industry of tomorrow
Viventris develops smart and flexible production solutions that help companies improve their processes and make them future-proof. We specialize in visual inspection and assembly, with applications in consumer products, high-tech systems, and healthcare, among others.
Our strength lies in combining generic building blocks with customer-specific customization. This allows us to realize efficient and reliable solutions for the production of compact and often complex products. We work pragmatically and closely with our clients to achieve results quickly and make a real impact on the shop floor.
Within our focus area of assembly, we automate manual processes and complex operations, leading to increased product quality and improved efficiency. Within our inspection expertise, we develop advanced vision systems utilizing both rule-based algorithms and AI. This enables accurate product inspection and reliable defect detection.
With our solutions, we contribute to smarter production processes and consistent product quality.
Vision workflows in MLOps for inspection systems
Inspection is a core competency within Viventris, with applications such as defect detection, metrology, surface inspection, and print inspection. Projects utilize both commercial tools (such as HALCON) and open-source vision solutions.
To further scale these solutions and make them deployable more quickly for new applications, there is a need for more standardization in vision workflows and a stronger setup of MLOps processes. This includes structured management of data, models, and deployment within different projects.
In this internship, you will work on further professionalizing and standardizing vision workflows, with a strong focus on MLOps and open-source tooling.
Goal
Developing a standardized vision workflow and MLOps approach for inspection applications, focused on reusability, scalability, and efficient development.
What are you going to do?
You start by analyzing existing vision and AI projects and mapping out how processes such as annotation, training, and deployment are currently set up.
Next, you will work on a standardized workflow based on open-source tooling, which brings together the following components:
- image data annotation;
- model training;
- MLOps aspects such as data versioning, model versioning, and performance tracking;
- Deployment and inference of models.
You research and compare different tools and frameworks, and make informed choices based on this.
The focus is on establishing a reproducible and scalable pipeline that can be applied across multiple projects and integrate with existing software platforms.
To make the approach tangible, you will work it out in a demonstrator, for example, a print inspection setup in which the entire workflow is followed from data to inference.
Result
A functioning and documented vision workflow with an integrated MLOps approach, including a demonstrator and guidelines for application in future projects.
Required skills
- Experience with Python
- Interest in industrial inspection applications
- Interest in computer vision and machine learning
- Adv: Experience with MLOps tooling (e.g., experiment tracking or data versioning)
- Experience with deep learning frameworks (PyTorch, TensorFlow)