Data Scientist in high-tech scale-up

Luo Automation B.V; Delft; On-site; EUR 65,000 - 90,000; Full time

Job summary

Luo Automation B.V. is seeking a Data Scientist to advance ML models for high‑tech pharmaceutical inspection systems. You’ll develop and optimize models, manage data strategies, and work closely with Software, Quality Support and Machine Vision teams to deliver real‑world impact.

You should have a HBO data science/AI background, hands‑on Python and ML framework experience, and a focus on image data. This is an on‑site role at our Delft office, with collaboration across cross‑functional teams.

Qualifications

• Bachelor HBO in data science, AI, Computer Science, or a related field.
• Hands-on experience with Python and machine learning frameworks such as PyTorch or TensorFlow.
• Experience analyzing image data; medical or industrial datasets is a plus.
• Able to own the full pipeline from data collection to experiments.

Responsibilities

• Develop and optimize ML models for real-world pharma applications.
• Collaborate with Software, Quality Support, and Machine Vision teams to deliver impact.
• Translate research into concrete improvements meeting regulatory needs.

Skills

Python

Machine Learning

Data Analysis

English fluency

Dutch (nice to have)

Education

Bachelor HBO in data science/AI/CS or related

Tools

PyTorch

TensorFlow

Job description

Now Hiring: Data Scientist in high-tech scale-up

Location: Delft, on-site

At Luo, we build machines that automate quality inspections of medicine. In many pharmaceutical environments, people still spend hours manually checking products for particles, cosmetic defects, and packaging integrity issues. It’s tedious, and humans make mistakes. Our machines do this job more accurately, faster, and more consistently.

The brains of our machines are the detection models: algorithms that decide in real time whether a product passes inspection. We’ve proven the technology works. Now we’re scaling, which means more clients, more applications, and higher accuracy. For that, we need someone who builds, improves, and pushes these models to the next level.

Our office is located at YES!Delft, surrounded by dozens of other tech start-ups and scale-ups, right next to TU Delft. Much of the work happens hands‑on, with our machines, test setups, and the teams around them.

The Role

We are looking for a Data Scientist who enjoys developing and optimizing machine learning models for real‑world applications. Not just running models and reporting results, but understanding what has impact and executing what must be done to make it better. Most of the time, that is more than just re‑training.

You’ll work on detection models for particle contamination, cosmetic defects, and closure integrity problems. You stay on top of the latest developments in the field, translate them into concrete improvements, and ensure our models meet the strict requirements of the pharmaceutical industry. You work closely with the Software, Quality Support, and Machine Vision teams, who support you as you grow in the role.

What you’ll work on

• Staying up to date: Following the latest developments in detection models and AI, assessing what’s relevant for our custom models, and translating this into concrete research actions.
• Building and improving detection models: Adapting existing models to new clients, improving detection, and qualifying model performance offline.
• Orchestrating data strategy: Managing and executing dataset annotation for model training and defining which data needs to be available to build and improve the best detection models.

What we are looking for

• Education: A bachelor’s (HBO) in data science, AI, Computer Science, or a related field.
• Strong ML foundation: You have hands‑on experience with Python and machine learning frameworks such as PyTorch or TensorFlow, gained at work or during your studies, graduation project, or internship.
• Data analysis: You have a strong focus on analyzing image data, understanding patterns, and extracting meaningful insights to improve model performance. Experience with medical or industrial datasets is a plus.
• Hands‑on experience: You are comfortable owning the full pipeline, including data collection, annotation, and running experiments independently, or you’re eager to learn this quickly.
• Analytical and communicative: You can explain complex findings clearly to non‑technical people and know how to tailor your message to your audience.
• Languages: Professional fluency in English, both written and spoken, is required. Dutch language skills are considered a plus.
• Eye for quality: You recognize that accuracy is critical in the pharmaceutical industry and understand the real‑world impact of your models, especially when it comes to patient safety.
• Direct and transparent: You communicate openly about challenges. If a model underperforms or data is insufficient, you raise your concerns early so we can solve them together.
• Office-based work is required, allowing for fast communication and close collaboration with your team.
• Dutch/EER working permit

What it’s like to work here

• Real impact: Your models run in real machines at real client sites. When you improve something, you directly see the effect!
• Flat structure, short lines: Good idea? Go for it. Our structure is simple to encourage involvement, creativity, and innovation.
• Interesting challenges: Detection problems in the pharmaceutical industry are complex and demanding. You’ll work at the edge of technology, where quality standards are high.
• A solid team: You work alongside engineering, operational, and commercial teams who all share the same goal. Cooperation and teamwork are key to building and delivering machines that actually work.