Nathan Wilkinson Email & Phone Number

Amsterdam, North Holland, NL

Increasing the scalability and reliability of the Weaviate OSS distributed vector database.

Salt Lake City, UT, US

Designing and building a data platform to store and process >50 billion rows of biological data.Assembling efficient training and inference infrastructure to predict protein/molecule binding.Created and released one of the largest public protein/chemical binding datasets in the world (https://www.kaggle.com/competitions/leash-BELKA/overview).

South San Francisco, California, US

Build and design backend systems and data pipelines underlying the payments infrastructure of the internet.

Develop computational pipeline that processes many terabytes of biological images weekly.Engineer data analysis tools to make it easy and fast for data scientists to run machine learning and deep learning models.

Developing cancer diagnosis pipeline on AWS in Python to identify somatic DNA mutations.Designing pipeline to process terabytes of data in parallel.Classifying structural variants from DNA sequence data with machine learning.

Salt Lake City, Utah, US

Deployed Go application on gcloud, K8, and Docker to consolidate laboratory robotics log data from Splunk server into Postgres database for reconciliation with experiment plans.Created and optimized scalable Postgres view with millions of entries to instantly find robotics errors that were previously impossible to detect.Designed Go microservice to manage.

Used Python, R, and machine learning to develop a predictive regression model (support vector regression) for crack propagation in three dimensions.Feature extraction from billions of data points increased model accuracy by 400%.Wrote Python program to genotype structural variants from DNA sequencing data.

Salt Lake City, UT, US

Contributed over 5000 lines of production code in Java stack to implement a new feature that integrated geospatial data into the existing codebase.