AI Research Engineer – Reinforcement Learning in Portugal.
This role sits at the forefront of applied AI research, focusing on advancing reinforcement learning systems that power next-generation intelligent models.
You will design and optimize algorithms that improve decision-making, adaptability, and performance across complex, real-world environments.
Working in a highly research-driven and experimentation-heavy setting, you will contribute to both foundational RL innovations and production-grade implementations.
The position spans work on efficient models for constrained hardware as well as large-scale multimodal systems integrating text, image, and audio.
You will play a key role in building simulation environments, refining training pipelines, and enhancing policy performance.
This is an opportunity to directly shape cutting-edge AI systems deployed at global scale.
Accountabilities
- Design and implement advanced reinforcement learning algorithms to improve decision-making, policy optimization, and system performance across simulated and real-world environments
- Run controlled experiments, track performance metrics, evaluate outcomes against benchmarks, and iterate on model improvements through empirical analysis
- Develop and curate high-quality simulation environments and training datasets aligned with domain-specific requirements and learning objectives
- Debug and optimise RL pipelines, addressing challenges such as exploration strategy, reward stability, sample efficiency, and training convergence
- Collaborate with engineering and research teams to integrate RL agents into production systems and ensure measurable real-world performance gains
- Define evaluation frameworks and continuously monitor deployed systems to support robustness, scalability, and domain adaptation
Requirements
- Advanced degree in Computer Science, Machine Learning, or related field;
PhD preferred with strong academic research background and publications in top-tier conferences
- Proven experience running large-scale reinforcement learning projects, including modern online RL techniques such as policy optimisation methods and actor-critic frameworks
- Deep understanding of reinforcement learning theory and practice, including policy gradients, exploration-exploitation trade-offs, and optimisation strategies for stability and efficiency
- Strong hands-on expertise with PyTorch and RL frameworks, including building full pipelines from simulation to training and deployment
- Demonstrated ability to solve complex RL challenges such as sample inefficiency, reward noise, and training instability through empirical and algorithmic innovation
- Strong analytical mindset with ability to design robust experiments, interpret results, and continuously improve model performance
Benefits
- Fully remote work environment with global team collaboration
- Opportunity to work on cutting-edge AI and reinforcement learning research at scale
- High-impact role influencing production-level AI systems and real-world applications
- Competitive compensation aligned with experience and expertise
- Exposure to advanced research, multimodal AI systems, and state-of-the-art infrastructure
- Flexible working culture supporting autonomy and innovation
#J-*****-Ljbffr