Roman Nozdrin
Machine Learning Engineer @Navan
Signup · Get unlimited contacts
WORK HISTORY
Machine Learning Engineer @Navan
Berlin, DE
As ML Engineer in Navan in a cross-functional team of 2 I- Shipped batch and real-time models to production end-to-end: collected requirements, created AWS infrastructure, trained and deployed models, set up monitoring and alerting in new relic. Designed and supervised A/B tests, implemented A/B routing logic on inference side- Built and maintained a python library for ML pipelines orchestration, experiment tracking, feature store and model registry management on top of AWS Sagemaker and integrating with Airflow and Snowflake- Wrote terraform modules to streamline deployment of real-time ML models, promote IaC reusability and compliance with SRE standards- Experimented with new ranking models using Ray and Sagemaker Experiments- Maintained legacy ML services and related ETL jobs.
EDUCATION
Saint Petersburg State University
Bachelor's degree, Applied Mathematics and Computer Science
ABOUT ROMAN NOZDRIN
I\'m a machine learning professional with a strong focus on measurable impact. I thrive on solving complex problems involving large-scale and heterogeneous data. I work proactively and take ownership across the entire lifecycle: from identifying high-leverage opportunities and gathering requirements, to building, deploying, and monitoring robust machine learning systems in production.TLDR; I make ML models go
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.