What We Do

You have a hard data problem and need someone who can own it end-to-end: from cleaning the mess to delivering an answer you can actually use. Full Rank has helped clients build entity resolution models, agentic RAG solutions, Bayesian cohort analyses, emissions visualizations, and more. Reach out to get started.

Recent Work

Startups are Sleeping on Responsible AI (Featured on GeekWire)

An analysis of the GeekWire 200 shows that while most startups market their AI, few explain how it works – or why they can be trusted. 73% of B2B startups are talking AI, but their hype outpaces their responsibility.

Read more

Building Responsible AI at Textio

Ryan developed statistical methods for bias evaluation, and coauthored the Building Responsible AI at Textio whitepaper.

Read the whitepaper

Interview: Bias in Data Science

Ryan Sloan spoke with Information Week about sources of bias in data science, and shared bias mitigation strategies.

Read the interview

High Performers get the Lowest-Quality Feedback

Ryan partnered with Kieran Snyder on an analysis of performance reviews and performance ratings across multiple companies.

Read the report

How We Work

We are not currently accepting clients. Full Rank works on both per-project and fractional retainer-based engagements. Project specifics are tailored to your needs and scope. We'll start with a discovery call to discuss the problem space. If it seems like a fit, we'll share a proposed project plan with a quote.

Ryan Sloan, Founder

Ryan Sloan's record in data science, product management, and strategy was built at companies large and small. His career spans 15 years at Microsoft, Code.org, Textio, and Syndio. Past colleagues have described him as having "a knack for presenting insights in a super crisp and clear way", "a grab-a-shovel mentality", and "strong teacher energy (in the best way)". Ryan's undergraduate and graduate studies were focused on computer science, statistics, and machine learning at Georgia Tech. Connect with Ryan on LinkedIn

Why call it "full rank"?

In linear algebra, a matrix with full rank is a special kind of matrix that always has a unique solution. Think of it as a maximum information-density matrix: all signal, no noise.