AI Research Engineer Intern en Instadeep

About Applied AI:The Applied AI team at InstaDeep creates optimization solutions for large scale real-world industrial problems. Each solution is built to be ready for full production use by our clients and it is the Applied AI team that owns all stages of this lifecycle: Within Applied AI, Research Engineering builds the AI optimization systems, MLOps sets up the training and inference infrastructure and the Software teams build graphical user interfaces and API servers. These teams work together to provide the end-to-end capabilities required to create scalable and high-performing optimization tools for the end user, all done 100% within InstaDeep. Our work spreads across many industries - from Energy to Logistics to Production Optimization, there are plenty of interesting topics!About the role:As an AI Research Engineer Intern, you will be responsible for developing innovative solutions to complex, real-world challenges. This role involves leveraging a broad spectrum of advanced AI methodologies, including but not limited to deep reinforcement learning, graph neural networks, and quality-diversity algorithms. You will actively engage in the research, development, and optimization of novel approaches, systematically evaluating and refining strategies to deliver effective, state-of-the-art solutions grounded in the latest advancements in artificial intelligence.This job posting groups several internship topics.

Responsibiliites:

Design, implement, and evaluate reinforcement learning algorithms and benchmarking experiments to address complex decision-making or control problems.

Leverage modern deep learning frameworks and classical optimization techniques to improve agent performance, stability, and generalization.

Analyze and validate model behavior through rigorous experimentation and statistical testing to ensure robustness and reproducibility.

Contribute to long-term research in AI-driven optimization, planning, and autonomous systems by developing novel RL methods or improving existing architectures.

Report results clearly and efficiently in written and verbal form, both internally and externally.

Write clean, maintainable, and modular code with thorough documentation.

Requirements:

Currently in the last year of a Master’s or PhD program in Computer Science, Applied Mathematics, or a closely related discipline.

Comprehensive understanding of deep learning principles, with solid foundations in reinforcement learning, and associated areas of research.

Proficiency in Python, with hands-on experience using leading deep learning frameworks such as PyTorch, TensorFlow, and JAX.

Demonstrated ability to work collaboratively within multidisciplinary research and engineering teams.

Exceptional communication skills and fluency in English, both written and verbal.

Solicitar ahora