Applied Scientist, Orc Science

PhD, or a Master's degree and experience in CS, CE, ML or related field
- Experience in patents or publications at top-tier peer-reviewed conferences or journals
- Experience programming in Java, C++, Python or related language
- Experience in any of the following areas: algorithms and data structures, parsing, numerical optimization, data mining, parallel and distributed computing, high-performance computing

Are you ready to make a global impact with your expertise in Machine Learning, Statistics, and Multimodality? Amazon is looking for an innovative and driven Applied Scientist to join our team and help shape the future of cutting-edge technology. In this role, you will design, develop, and deploy advanced machine-learning models to tackle some of the most complex and meaningful challenges in the digital world. From automating compliance processes to building systems capable of intelligent, autonomous classification, your work will directly enhance the experience of millions of customers on the largest online retail platform.
This is your opportunity to collaborate with world-class scientists and engineers, advancing innovation in machine learning, natural language processing, and statistical modeling. Your contributions will extend beyond technology—transforming research into scalable solutions that redefine customer experience and operational efficiency. Whether publishing scientific papers, developing patents, or presenting your work to stakeholders, you’ll have a platform to showcase your expertise while driving real-world impact.
Key Responsibilities
Conduct pioneering research in machine learning, statistics, and multimodal systems to create groundbreaking solutions for customer and operational challenges.
Analyze vast datasets to uncover actionable insights, design scalable algorithms, and seize opportunities for innovation.
Validate machine-learning models through rigorous statistical experiments involving millions of users.
Partner with software engineering teams to prototype and integrate successful models into global production systems.
Collaborate with a multidisciplinary team of applied scientists and engineers to push the boundaries of innovation.
About the Location
This role is based in Luxembourg, home to Amazon’s European headquarters. Nestled in the heart of Europe and bordered by France, Belgium, and Germany, Luxembourg is a vibrant, multicultural hub offering a high standard of living and easy access to major European cities. Known for its thriving financial and tech sectors, Luxembourg combines innovation with a rich cultural scene and stunning natural landscapes. Discover more about life in Luxembourg at Promote Luxembourg.
Key job responsibilities
Push the boundaries of machine learning by developing cutting-edge models to classify products on Amazon's website. Leverage multimodal approaches, transformers, and computer vision to create state-of-the-art solutions.
Dive into innovative research on ML techniques to overcome the challenge of meeting strict latency requirements for real-time online inference, shaping the future of seamless customer experiences
A day in the life
A Day in the Life of an Applied Scientist II at Amazon
As an Applied Scientist II at Amazon, each day is a blend of technical challenges, innovation, and collaboration. Here’s a glimpse into a typical day:
8:30 AM - Start the Day with Focus
9:00 AM - Team Stand-Up
Join the daily stand-up meeting with scientists, engineers, and stakeholders. Share updates on experiments, discuss challenges, and align on priorities. The collaborative environment ensures everyone is on the same page.
10:00 AM - Deep Dive into Research
Dive into ongoing research tasks. This could mean reading the latest papers on transformers or multimodal learning, brainstorming ways to integrate new ideas into Amazon’s systems, or designing experiments to test novel approaches.
12:00 PM - Lunch and Networking
Take a break to recharge. Amazon offers opportunities to connect with peers over lunch, whether it’s discussing industry trends or exploring ideas for cross-team collaboration.
1:00 PM - Model Development and Experimentation
Head into coding and experimentation mode. You might be refining a multimodal model for product classification, training transformers on massive datasets, or fine-tuning a computer vision model to improve prediction accuracy. Metrics are monitored closely to ensure models meet Amazon's high standards.
3:30 PM - Collaboration Time
Meet with engineers to optimize deployment strategies for models in production. Discussions often focus on meeting latency requirements for real-time inference without compromising accuracy. This synergy between science and engineering is key to delivering scalable solutions.
4:30 PM - Results Review and Insights
Analyze experimental results, document findings, and prepare for upcoming iterations. Insights gained here shape the next steps, whether it’s tweaking hyperparameters or exploring a new