pylon AI

Seamless integration of custom ML vision models with Basler’s ecosystem

Design and development:

07.2022 – now

Team size:

5

CHALLENGE

Basler, a leader in image processing and computer vision technology, partnered with us to develop a groundbreaking solution enabling their clients to seamlessly integrate custom machine learning models with Basler's hardware and software.

Basler's clients were facing increasingly complex needs in production line monitoring, video anonymization, and quick navigation through months of data - use cases that often required custom machine learning models. As a company providing advanced camera suites to hundreds of businesses worldwide, Basler recognized an opportunity to enhance their offerings with Machine Vision capabilities.

  • Determining optimal hardware and software configurations for customer ML models.
  • Delivering ready-to-use installation bundles tailored to individual hardware setups.
  • Maintaining a rapid development pace to meet tight deadlines.
  • Presenting an MVP at Stuttgart's Vision Show, a premier industry event.

Building on a successful prior collaboration with Basler’s Head of R&D, we were entrusted to tackle these challenges. Our expertise in initiating and managing innovative software projects positioned us as the ideal partner for this venture.

Key challenges

SOLUTION

Strategic Planning Meets Agile Execution

Our approach was rooted in collaboration, transparency, and precision. By combining technical expertise with agile processes, we delivered both an MVP and ongoing enhancements for the application. Our support continues to be integral to the project’s development.

Key Steps in the Process

1

We assembled a specialized team of software developers, DevOps specialists, UX/UI designer, and business analysts in less than a month - significantly faster than in-house recruitment.

2

We arranged a series of workshops and regular meetings with Basler specialists to gather and refine both business and technical requirements. This was also the stage at which the initial Figma mockups and Azure Cloud architecture concepts were developed.

4

For the next several weeks, we developed a scalable process supporting both immediate development needs and future team growth.

3

With them in hand, we selected technologies precisely matched to project requirements, including microservices and containers.

5

Our team created interactive mockups that Basler could showcase at the Vision Show while development continued.

6

We deployed the application with careful attention to Basler's monitoring and security requirements.

7

Following the deployment, we transitioned into continuous development, adding new features such as statistical analysis, data visualization, classification, and segmentation use cases.

/>
/>
1

We assembled a specialized team of software developers, DevOps specialists, UX/UI designer, and business analysts in less than a month - significantly faster than in-house recruitment.

2

We arranged a series of workshops and regular meetings with Basler specialists to gather and refine both business and technical requirements. This was also the stage at which the initial Figma mockups and Azure Cloud architecture concepts were developed.

3

With them in hand, we selected technologies precisely matched to project requirements, including microservices and containers.

4

For the next several weeks, we developed a scalable process supporting both immediate development needs and future team growth.

5

Our team created interactive mockups that Basler could showcase at the Vision Show while development continued.

6

We deployed the application with careful attention to Basler's monitoring and security requirements.

7

Following the deployment, we transitioned into continuous development, adding new features such as statistical analysis, data visualization, classification, and segmentation use cases.

Model Verification

Pylon AI enables clients to test their models on various hardware platforms to find the best price-to-performance ratio.

Custom Installation Bundles

Users can upload their models and datasets to generate installation bundles optimized for Basler hardware.

Microservice Architecture

Designed for fault tolerance, this architecture minimizes dependencies between application components, enhancing reliability.

Azure Cloud Integration

Leveraging Azure services (e.g., storage, queues, databases) accelerated development while reducing implementation costs.

RESULTS

Basler customers harness the power of Machine Learning.

Pylon AI empowers manufacturers by simplifying ML model deployment. Our long-term collaboration helped Basler solidify its position as a trusted partner for Mittelstand companies navigating Industrie 4.0.

20

application screens

3

months of Design Phase

6

months of MVP development

18+

months of support and regular application development

6

months of MVP development

3

months of Design Phase

20

application screens

18+

months of support and regular application development

Technology stack

No items found.

The enhancement of machine learning and computer vision is what sets this project apart. On top of that, a successful and long-term collaboration with Basler shows that Codya is capable of working hand in hand with large, international enterprises seeking to build their competitive advantage through technology.

Grzegorz Lewiński

Software Developer at Codya

Other case studies

/ need a hand?

You have a challenge. We’re there to solve it.
Let’s grab a (virtual) coffee and discuss your project.

Contact us