Miguel Varela Ramos

Miguel Varela Ramos has a diverse work experience in the field of software engineering and machine learning. Miguel is currently working at Cohere as a Senior Software Engineer - Infrastructure, a position they have held since February 2022. Prior to that, Miguel worked as a Software Engineer - Infrastructure at the same company from October 2021 to February 2022.

Before joining Cohere, Miguel worked at Cortex Labs from December 2020 to October 2021. Miguel held the role of Software Engineer - ML Infrastructure and was involved in the development of Cortex's Kubernetes operators, specifically focusing on managing Custom Resource Definitions (CRDs) for different application types. Cortex Labs was acquired by Databricks.

From September 2019 to November 2020, Miguel worked at Sensity as a Senior Machine Learning Engineer. Sensity AI, formerly known as Deeptrace, focused on using deep learning and computer vision to detect and monitor AI-generated media. While at Sensity, Miguel designed and implemented an event-driven, asynchronous, and microservices-based API to process requests for analyzing videos and images.

Prior to joining Sensity, Miguel worked at BrainCreators from May 2018 to August 2019 as a Machine Learning Engineer. During their time there, they worked on various projects, including the development of speech recognition models for English and Dutch, the creation of an unsupervised machine learning pipeline for manual labeling of network anomalies, and the development of an object detection pipeline. Miguel also made contributions to several open-source projects, including those by Facebook Research and Uber.

Before BrainCreators, Miguel worked at Unbabel from January 2017 to April 2018 as a Machine Learning Scientist. Unbabel is an AI start-up that combines machine translation with human post-editors to ensure the quality of translated text. While at Unbabel, Miguel was part of the team that created a framework-agnostic codebase for training and inference of any machine learning model. Miguel also implemented a quality estimation framework in PyTorch for post-edited text.

In addition to their industry experience, Miguel has also worked as a Research Intern at Carnegie Mellon University in late 2016, where they worked on voice conversion and achieved state-of-the-art results with a fraction of the required dataset. Prior to that, they worked as a Master Student Researcher at INESC-ID from September 2015 to October 2016.

Miguel Varela Ramos obtained a Master of Science (MSc) degree in Electrical Engineering (Systems, Decision and Control) from Instituto Superior Técnico. Miguel pursued this degree from 2010 to 2016. Additionally, they participated in an Erasmus Exchange program at Delft University of Technology, where they also earned a Master of Science (MSc) degree in Electrical Engineering in 2013 and 2014. Miguel Varela Ramos also possesses an IELTS certification, obtained from the British Council, although the specific date of obtaining this certification is unavailable.