Sandra Meneses is a Senior ML Engineer based in Berlin with 11 years of experience designing and shipping end-to-end machine learning systems. She combines hands-on model development and data engineering with production deployment expertise, having built recommender systems, serverless data pipelines, and serving infrastructure using tools like PyTorch, DGL, Snowflake, dbt, and AWS (SageMaker, Lambda, ECS Fargate). At Tandem she implemented reciprocal recommenders and metrics-driven evaluation; at Mercedes-Benz.io she turned ML prototypes into backend and mobile integrations while collaborating closely with product and DevOps. An active contributor to NLP tooling, she improved the popular flairNLP project by refining STACKOVERFLOW_NER preprocessing, entity mapping, and corpus logging to boost dataset usability. Currently a Senior ML Engineer at PyMC Labs and member of AI Guild, she blends production-grade engineering with mentorship and teaching experience from Data Science Retreat and academia. Her background in industrial engineering and a master's in data science gives her a pragmatic systems perspective that ties modelling choices to operational impact.
11 years of coding experience
5 years of employment as a software developer
Specialization in Production Engineering Engineering/Industrial Management, Specialization in Production Engineering Engineering/Industrial Management at Universidad Distrital Francisco José de Caldas
Master's degree Data science, Master's degree Data science at Hochschule für Technik und Wirtschaft Berlin
Bachelor's degree Industrial Engineering, Bachelor's degree Industrial Engineering at Unversity of Pamplona
A very simple framework for state-of-the-art Natural Language Processing (NLP)
Role in this project:
ML Engineer
Contributions:5 commits, 2 PRs, 1 comment in 14 days
Contributions summary:Sandra focused on enhancing the `flairnlp/flair` repository, specifically related to the STACKOVERFLOW_NER dataset. They implemented an entity mapping, added data cleaning steps, integrated logging for corpus summaries, and introduced a "banned sentences" parameter. These changes improved the dataset's usability by refining the NER process, cleaning the data, and enhancing the corpus handling and reporting.
Contributions:98 commits, 3 pushes in 1 year 1 month
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.