Sustaining ML in the enterprise
summary
Survey of 11,000 respondents finds many orgs already in production with machine learning and a third using serverless options while ML practitioners rely on Spark Kafka TensorFlow and PyTorch for pipelines. Practical wins include MLflow adopted by 200+ companies for experiment tracking and claims of upcoming hardware yielding 15–40x training speedups. Production needs center on data integration and cleaning, data catalogs and lineage (Data book Marquez DVC Delta Lake), and model governance plus monitoring and provenance dashboards.
AI-written summary · about
chapters (9)
- 0:00Strata DATA CONFERENCE
- 1:08Google Machine Translation
- 2:08What ML means for software development.
- 6:54Which solution is your organization currently building or evaluating?
- 11:40Open source platform to manage ML development
- 12:42Machine Learning in Practice
- 14:18Model Governance & Model Ops
- 15:50Managing Risk in Machine Learning
- 16:34GDPR: May 2018 - present
related talks
34mScaling ML in the Traditional Enterprise
41mInfrastructure for ML Applications
47mMachine Learning in Production: An Intro to MLOps.
41mTHE STATE OF MLOPS - MACHINE LEARNING IN PRODUCTION AT ENTERPRISE SCALE
21mHow The Application Of Machine Learning Is Changing The Way Companies Use Data
11mRecent trends in data and machine learning technologies
13mMLOps: From AI Research to Production-Ready Systems
27mA Look Inside Applied Machine Learning