Miguel Angel Rubio

Machine Learning Engineer @Amazon

Cambridge, GB

As a Machine Learning Engineer at Amazon AGI, I focus on enhancing large-scale AI systems to improve user experiences across Alexa-enabled devices.​- Developed and fine-tuned large language models (LLMs) to enhance customer interactions for over 500 million Alexa devices- Implemented a real-time context retrieval system, increasing LLM accuracy and relevancy, leading to a 4% rise in successful query responses- Led the migration of fine-tuning pipelines from NVIDIA GPUs to AWS Trainium instances, improving training efficiency by 5% and reducing costs- Built and maintained internal tooling utilized by over internal customers, streamlining data preparation, fine-tuning, and evaluation processes for LLMs- Designed a scalable data collection service processing 365 petabytes of data weekly, reducing operational costs by 90% without compromising speed- Integrated new machine translation providers into existing systems, enabling support for additional languages and facilitating Alexa\'s expansion into new markets.​

Universitat Politècnica de València (UPV)

Bachelor's Degree in Business Administration and Management, Project management

Universitat Politècnica de València (UPV)

Bachelor’s degree in computer science, Project management