Pamela Wu Email & Phone Number

Ottawa, ON, CA

New York, NY, US

Conducting research on viability of automatic training data generation using Snorkel and designing a system for entity resolution validation and manual label checking.Designed and implemented several popular team processes, including a brand new set of data science interviews, collaborated on a job packet and set of interviews for our new data analyst.

New York, NY, US

Using a combination of a web scraping strategy, a training data acquisition strategy, a robust team culture around data evaluation and QA, and ML model architecture design, we now predict 70+ granular industries at 80% precision or above for each industry. Our business closure attribute has 90% precision with respect to our online sources. Our coverage.

New York, NY, US

Led the data science team that was responsible for improving our company search models and developing initial SMB attributes, such as verification and industry prediction. We increased match rates on a key client benchmark from 66% to 78%, with an increase in precision from ~85% to greater than 90%, through a combination of Elasticsearch query.

New York, NY, US

Created key algorithmic optimizations in Scala-Spark that allowed the entity resolution pipeline to handle the change in scale when the initial prototype was scaling from California companies to all US, a greater than 20X increase in volume, and these optimizations are still running in production today.

New York, NY, US

Used high-throughput sequencing tools and statistical analysis to uncover novel long non-coding RNA targets for melanoma disruption and performed differential proteomic analysis for finding potential protein structures that underlie the creation of blood vessels feeding pediatric optic gliomas. I also worked with the cBioPortal team at MSK to create.

New York, NY, US

Worked on a project to automate integrating mass spectrometry proteomics data from both the CPTAC official platform and MaxQuant output into the platform on cBioPortal, an open source cancer genomics data exploration portal. This extended into updating their OncoPrint heatmap feature to display genomics and proteomics expression data. We published a paper.

New York, US

Led a team of consultants for Greenhouse where we did a competitive analysis for a product line expansion that they were considering. Consulted for Dropbox on doing research on an industry vertical they were considering expanding into.

Ottawa, On, CA

Data processing and scripting in MATLAB while persuing projects in cell mechanics and wound healing assays.

Honours project was on the localization of certain transcription factors in the fish forebrain that are also found in the human neocortex. The idea is that where these genetic signals are located can shed light into forebrain evolution.

Assisted in research in fruit fly genetics and evolution, where I also pursued an independent project on whether sperm depletion of high condition males could explain females choosing lower condition males.

Phd, Systems And Computational Biomedicine

Bachelor'S Degree, Biology/Biological Sciences, General