ML Engineer

Company: Konica Minolta Video Solution Services Business Unit

Location: [Hybrid – Brno]

Employment Type: Full-time

Position: Machine Learning Engineer (MLE)

About us

At the forefront of industrial innovation, we blend machine vision camera technology with advanced

AI algorithms to build products that redefine visual quality inspection and elevate workplace safety.

We are an international team based in Brno, working in hybrid mode (50/50). We meet in the office

for team ceremonies, design sessions, and planning meetings. Outside of work, we enjoy getting

together for team-building activities and social events

What we build

Our core product leverages state-of-the-art image processing techniques to detect, count, and inspect

items in real time.

Imagine an intelligent system monitoring a production line, identifying defects as they happen, and

even triggering hardware actions such as halting production or separating defective products from the

good ones. This is without human intervention. It's AI-driven automation with real-world impact,

helping clients save costs and raise quality standards.

We don't stop at quality checks. Our (no-code) AI platform also safeguards employee safety, health,

and well-being. With products like Forxai Mirror, we ensure workers are equipped with the correct

Personal Protective Equipment (PPE), keeping them compliant with safety regulations automatically

and reliably.

About the role

We are seeking an ML Engineer with hands-on experience in deep learning, model deployment, and

real-world applications of computer vision systems. You will work closely with our development team

and customers to design, optimize, and deploy ML models tailored to specific use cases. This is a hybrid

role that combines core ML development with engineering support and insight into production

behavior.

Key responsibilities

• Develop, train, and fine-tuned computer vision models using PyTorch and Python.

• Convert and optimize models using ONNX, TensorRT, and deploy with Triton Inference

Server.

• Diagnose and improve underperforming models through error analysis and iterative

refinement.

• Work on customer-specific projects, delivering vision solutions adapted to varying real-world

conditions and constraints.

• Collaborate with full-stack developers, DevOps, and project leads to integrate ML systems

into production.

• Contribute to efficient model packaging, deployment pipelines, and observability tools.Requirements

• Proven experience building and deploying computer vision models in production.

• Proficiency in Python, PyTorch, and common vision libraries (OpenCV, torchvision).

• Experience with model optimization and serving: ONNX, TensorRT, Triton.

• Strong debugging skills and the ability to evaluate performance bottlenecks in models and

pipelines.

• Experience with CI/CD for ML or MLOps tools is a plus.

Nice to have

• Experience with Azure ML, Azure Blob Storage, or other cloud-based ML pipelines.

• Knowledge of real-time or edge deployment.

• Understanding of containerization (Docker) and inference monitoring tools.

What we value

We are looking for skilled engineer who promote collaboration, knowledge sharing and enhancing

expertise across the team. We value the drive for ensuring customer satisfaction and delivering quality

products that meet business requirements. Good written and verbal communication skills and

proactive approach to problem solving and improving product.

If you're passionate about machine learning, computer vision, and applying AI to solve real-world

challenges, we'd love to hear from you.

What we offer:

• Flexible working hours and home office

• 5 weeks of holidays and 5 sick days

• Bonuses

• Self-Development budget CZK)

• Mobile phone of your choice

• Meal Voucher tariff

• Multisport card

• Language lessons of English, German, Italian and Japanese

• Refreshments in the office and occasional catering during team meetings