Manchester, England, United Kingdom
I am currently a PhD candidate at Lancaster University, where my research focuses on improving and optimizing environmental monitoring using cooperative swarms of UAVs and UGVs. My work combines deep reinforcement learning, Gaussian process regression, and robotic systems to design intelligent and adaptive sampling strategies for complex environments. Alongside my doctoral research, I have professional experience as a Research Associate, working on machine translation, speech separation, speech enhancement, and voice activity detection. This has given me a strong foundation in applied machine learning for both speech and robotics domains. I am passionate about leveraging AI, robotics, and signal processing to solve real-world challenges that directly benefit people and society, and I am exploring early-stage commercialization of my research to translate swarm robotics technology into practical environmental solutions. Key Skills: Deep Reinforcement Learning Multi Agent Deep Reinforcement Learning Machine Learning and Artificial Intelligence Gaussian Process Regression & Bayesian Optimization Robot Operating System (ROS) & multi-agent coordination PyTorch & TensorFlow (deep learning frameworks) MATLAB & Python (scientific computing and prototyping) Speech Processing (speech separation, enhancement, VAD, machine translation) Data Engineering & Dataset Design (synthetic mixtures, noisy environments) Research Communication (academic publications, presentations, stakeholder engagement)
In partnership with Collaboraite, I contributed to advancing multi-lingual speech to text translation technologies by working with large encoder-decoder models. My role involved speech preprocessing tasks such as speech separation, enhancement and voice activity detection as well as studying hallucinations to understand their causes and develop effective mitigation strategies. This experience honed my analytical skills and deepened my knowledge of speech processing and deep learning.