Joseph Harder, Developer in Oakland, CA, United States
Joseph is available for hire
Hire Joseph

Joseph Harder

Verified Expert  in Engineering

Data Scientist and Developer

Location
Oakland, CA, United States
Toptal Member Since
November 21, 2022

Joseph is a data scientist with four years of experience in the dental, marketing, and government industries. He's also built standardized data storytelling tools. Most recently, he became the domain expert on operations data for Spin, an electric scooter company, a role that involves labor efficiency, scooter placement and use patterns, and geospatial data analysis. Joseph adds value by verifying and assessing needs and creating innovative quantitative solutions to answer meaningful questions.

Portfolio

Spin
Python, PostgreSQL, Google Cloud Platform (GCP), Data Build Tool (dbt)...
Nacci
Python, Docker, JavaScript, Amazon Web Services (AWS), SQL, D3.js...
Columbia University
C, Python, Bash, Linux

Experience

Availability

Part-time

Preferred Environment

Linux, MacOS, Python, PostgreSQL, Google Sheets

The most amazing...

...thing I've delivered is a set of optimization simulations that helped increase profitability and labor efficiency of electric scooters.

Work Experience

Senior Data Scientist

2021 - PRESENT
Spin
  • Increased the electric scooter utilization rate by 8% by optimizing scooter deployment neighborhoods.
  • Helped reduce weekly theft of electric scooters by up to 50% in some operating markets by contributing with analysis and data visualization tools.
  • Ensured accurate self-serve analytics by maintaining and creating critical data models using dbt and LookML and dashboards with Looker and Google Sheets.
  • Planned and prioritized projects with stakeholders and answered business questions related to user behavior and growth with data deep dives and root cause analysis.
Technologies: Python, PostgreSQL, Google Cloud Platform (GCP), Data Build Tool (dbt), ETL Tools, Looker, Business Intelligence (BI)

Senior Data Scientist

2018 - 2021
Nacci
  • Generated a biweekly newsletter for a branch of the US government surfacing social media trends using NLP tools such as BERT and topic modeling.
  • Built and deployed an ML-powered tool that extracted video, audio, and textual insights to generate audience clusters for artists and labels using MLflow, OpenCV, and clustering.
  • Improved analyst workflows by designing KPI dashboards with Metabase and data extraction tools for a national dental staffing company.
Technologies: Python, Docker, JavaScript, Amazon Web Services (AWS), SQL, D3.js, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), GPT

Graduate Student Researcher

2012 - 2017
Columbia University
  • Developed C/C++ software suites for Brownian Dynamics simulations and Monte Carlo to study cell growth and non-equilibrium nanoparticle motility.
  • Used advanced statistical analysis methods, including time series analysis and clustering, to understand the behavior of biophysical systems.
  • Published work in high-impact journals, including a cover article published in The Journal of Chemical Physics called The Role of Particle Shape in Active Depletion. Also, presented work at conferences.
Technologies: C, Python, Bash, Linux

Fit to Print

http://fittoprint.xyz/
Built and deployed Fit to Print, an NLP classifier that could predict the original publication of a news article with up to 70% accuracy with the methodology featured in Building Machine Learning Powered Applications (Emmanuel Ameisen, O'Reilly Media, 2020).

Languages

Python, Bash, SQL, C, JavaScript

Platforms

Linux, Google Cloud Platform (GCP), Amazon Web Services (AWS), MacOS, Docker

Tools

Looker, Google Sheets

Paradigms

Business Intelligence (BI)

Storage

PostgreSQL

Other

Chemistry, Physics, Data Build Tool (dbt), ETL Tools, Natural Language Processing (NLP), Data Analytics, GPT, Generative Pre-trained Transformers (GPT), Simulations

Libraries/APIs

D3.js

2012 - 2017

PhD in Computational Chemistry

Columbia University - New York, NY, USA

2008 - 2012

Bachelor's Degree in Chemistry

UC Berkeley - Berkeley, CA, USA

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring