HomepageData + AI Summit 2022 Logo
Watch on demand

Designing Better MLOps Systems

On Demand

Type

  • Session

Format

  • Hybrid

Track

  • Data Science, Machine Learning and MLOps

Difficulty

  • Beginner

Room

  • Moscone South | Upper Mezzanine | 156

Duration

  • 35 min
Download session slides

Vue d'ensemble

Real-world data problems are becoming increasingly daunting to solve, as data volume grows and computing tools proliferate. Since 2018, Gartner has predicted that 85% of ML projects will fail and this trend will likely continue through 2022 as well. Nevertheless, in most cases, ML practitioners have the opportunity to avoid their projects from failing in the early phases.

In this talk, the speaker will borrow from her consultancy and hands-on implementation experience with cross-functional clients to share her takeaways in designing better ML systems. The talk will walk through common pitfalls to watch out for, relevant best practices in software engineering for ML, and technical anchors that make a robust system. This talk aims to empower the audience – beginner and experienced practitioners alike – with confidence in their ML project designs and help provide the big-picture design thinking framework for successful projects.

Session Speakers

Chengyin Eng

Databricks

Visionnez les temps forts du Data+AI Summit

Watch on demand