Your machine learning projects don’t only need to succeed and produce results, but it’s crucial to keep operations on time, on budget, and most importantly, on point. That’s why MLOps is how organizations keep their ML projects on the right path. But for many teams, you might not know where to start. Don’t worry, ODSC West has you covered this November 1st-3rd. Check out the following sessions and learn how MLops can be implemented and streamline your project. 

Full-stack Machine Learning for Data Scientists

Bridging the gap in the machine learning stack can be a daunting task. From data, computing, versioning, software architecture, model operations, orchestration, feature engineering, and model development, an ML team has many layers to get through. In this session, get an introduction to how to think about these layers, where you need to go, tools, workflow, and more. 

Democratizing Distributed Compute & Machine Learning: A Tour of Three Frameworks

What is Democratizing Machine Learning, and what are the differences in documentation, open-source governance, usability, and more? In this session, you’ll learn about Apache Spark, Ray, and Dask in this hands-on coding workshop. 

MLOps for Deep Learning

In this session, learn how MLOps can address the challenges of life-long retraining for Deep Learning models. By allowing your deployment pipeline to address these challenges, you can serve reliable AI predictions. This discussion will look at an open-source project that integrates unique drift detection and model retain algorithms. 

Extensible Hosted Jupyter Notebook Platform for Accelerating Data Insights

Learn how to utilize ad-hoc data analysis with LinkedIn’s Darwin. Discover how Jupyter notebooks allow users to increase their productive compact while enabling them to share their findings with a broader audience while empowering their data with libraries such as TensorFlow.

Orchestrating Data Assets instead of Tasks, with Dagster

It’s time to think outside of the “Task” box and learn how to keep your data assets up-to-date. In this session, learn how software-defined assets implemented in Dagster, can run the computations you require while providing an open-source data management system that allows users to trust their data.

Data Analytics at Scale: A Four-Legged Stool

Enterprise analytics requires effort as data integration takes center stage. Because of this, data science professionals report spending upwards of 80% of their time working on data preparation. In this session, join Michael Stonebraker, PhD, of MIT, as he discusses DBMS and how it makes data integration manageable with data lakes. 

Data Science Platforms are Bad

Learn how you can secure the flexibility and capabilities your team needs without being stuck in the business of finding the latest methods of libraries. In this session, learn how you can find the tools you need to allow your data team to focus on data work, not data platforms. 

ML Tools for Humans

Machine Learning models come with inherent challenges. How are teams able to keep their projects in line with the proper model training, preprocessing, expected predictions, and more? In this talk, discover how a standard kit for developer tools for Machine Learning practitioners and teams can remove many of these difficulties while providing them with the flexibility needed. 

Register for ODSC West here.

There you have it, ODSC West 2022’s MLOps sessions. A wide variety of topics are sure to answer the machine learning questions you have in mind. If you haven’t gotten your pass yet, then don’t let this opportunity slip away. Listen, learn, and connect with the experts who can help you supercharge your machine-learning operations. Register now!