Data Scientist - Benchmarking Analytics (Python and/or R)

I run a growing consulting business in Toronto, Canada, and I’m looking for a sharp, analytical, and resourceful Data Scientist to support an important project for a U.S.-based company that is a leader in measurement, experimentation, and benchmarking for software and digital instrumentation. This is an opportunity to work on one of the most advanced data-driven companies in the world, directly contributing to setting benchmarking standards for the entire software industry.

Please Note: This will be a 10 week project but may extend into the future based on business volume and your performance.

This role is ideal for someone who enjoys dissecting and modernizing legacy data pipelines, solving complex data problems, and structuring large-scale datasets to generate insights that shape industry standards. A key part of the role will involve understanding and rewriting a legacy Python script, documenting it in clear, reproducible terms, and developing a modern, interpretable version that aligns with best practices.

Key Responsibilities

Legacy Python Code Analysis & Refactoring
• Audit and understand an existing legacy Python script used for data benchmarking.
• Develop a modern, interpretable, and optimized version of the script.
• Document newscript in plain English for internal stakeholders.

Benchmarking & Experimentation Analytics
• Develop new benchmarking metrics for industry-specific insights.
• Segmentation: Group, filter, run statistical summaries to glean trends in metrics
• Visualization: Create simple and meaningful visualizations.

Statistical & Analytical Work
• Perform correlation, regression, and other statistical methods to derive meaningful insights.
• Ensure all analytics work is documented and reproducible.

Data Cleaning & Transformation
• Ensure data integrity by handling duplicates, missing data, structure, and incorrect type casting.

Data Visualization & Reporting
• Develop clear, concise visualizations to communicate insights effectively.

What You Need to Succeed
• Expert Python (pandas transformation) skills – experience refactoring, rewriting, and documenting Python scripts is critical - you need to be able to work with Python at the "speed of thought "
• Alternatively, Expert R (tidyverse) skills - you need to be able to work with R at the "speed of thought " for creating visualizations and analysis, but comfortable in Python
• Ability to explain a complex statistical model in plain English (C2 required)
• Proficiency with data visualization libraries (Matplotlib, Seaborn, Plotly, or similar).
• Curious: Ability to work independently and think critically when dealing with messy data.
Bonus
• Experience with data benchmarking, product analytics, or large-scale experimentation.
• Comfortable using LLMs (ChatGPT, Claude, Perplexity) for troubleshooting and enhancing productivity.

How to Apply

To be considered, you MUST answer these questions:
• Tell me about a time when you found an interesting insight using logistic regression or another form of statistical analysis.
• What library or tool do you personally find indispensable for your work, and why?

Max 2 paragraphs. Applications without these two questions will be ignored.

To ensure that you've read until the end, please attach or link a picture of a white tiger in your application :)