Join Hugo Bowne-Anderson and Ryan Stevens (Head of Applied Science at Ramp, formerly at Meta) for a fireside chat exploring the lifecycle of machine learning systems and how to tackle some of their biggest challenges.
Drawing from Ryan’s experience leading applied science teams at high-growth companies, this conversation dives into building scalable, maintainable, and impactful ML systems that deliver real business value.
Key Topics of Discussion::
Solving Reproducibility and Maintenance Challenges: How Ramp addresses critical issues like audit trails, model inheritance, and reproducibility with YAML-based pipelines and structured ML workflows.
The Lifecycle of ML Systems: From the first models to production-ready systems—what it takes to deploy, iterate, and maintain ML systems effectively.
The Evolution of ML Teams: How applied science teams grow from causal analysis to supporting machine learning engineers, and how roles evolve to meet organizational needs.
Experimentation in B2B Organizations: Why smaller datasets and resource constraints require prioritizing the right hypotheses and focusing on high-impact experiments.
Scaling for Business Value: Lessons from Ramp’s journey to tackle credit risk, fraud, growth, and product challenges with ML systems.
Insights from an Economist-Turned-ML Leader: Ryan’s personal journey from causal analysis and econometrics to leading applied science teams and how this perspective shapes his approach to machine learning.
This session is ideal for software engineers, ML practitioners, and technical leaders tackling challenges in reproducibility, scaling, and operationalizing ML systems. Whether you're deploying your first model or refining production workflows, you’ll leave with practical takeaways to guide your work.
*The event is hosted by our partner,
Outerbounds
Feb 19, 02:00 PM PST. Add to calendar: Google | Yahoo 
