Bay Area MLflow Meetup @ Databricks, San Francisco

Join us for an evening of tech-talks about MLflow and Machine Learning from Microsoft, Google, and Databricks.

Agenda:
6:00 - 6:30 pm: Food, Drinks, Beer & Wine
6:30 - 6:35 pm: Introductions
6:35 - 7:05 pm: Managing the Full Deployment Lifecycle of Models with the MLflow Model Registry
7:05 - 7:35 pm: MLflow on and inside Azure
7:35 - 8:05 pm: TensorFlow(X) Data Validation: Better ML through better data
8:05 - 8:30 pm: Networking

Talk 1 - Title: Managing the Full Deployment Lifecycle of Models with the MLflow Model Registry
Presenter: Mani Parkhe, Databricks
Abstract: MLflow is an open-source platform to manage the machine learning lifecycle, including experiment tracking, reproducible runs, and model packaging. In this talk, we provide an overview of the latest component of MLflow, the Model Registry, which serves as a collaborative hub where teams can share, discuss, use, inspect, and track the lineage of models. Model Registry was introduced in MLflow 1.4 and is in Private Preview on Databricks

With this addition, MLflow provides end-to-end management of the deployment lifecycle of models from experimentation to testing and production, complete with approval and governance workflows.

Bio: Mani Parkhe is an ML/AI Platform Engineer at Databricks, focusing on the open-source platform initiatives, enabling data discovery, training, experimenting, and deployment of ML models on the cloud. After spending 15 years building software for semiconductor chip CAD, Mani transitioned to building big data infrastructure, distributed systems and web services, and machine learning platforms. Prior to Databricks, he worked on data-intensive batch and stream processing problems at LinkedIn and Uber.

Talk 2 - Title: MLflow on and inside Azure
Presenter: Akshaya Annavajhala, Microsoft
Abstract:

During this presentation, after walking through a few ways to use MLflow on Azure directly, we'll cover how upcoming solutions from our group leverage MLflow for core functionality. BenchML is a new repository that aims to provide consumers of prebuilt ML endpoints visibility into the performance of each public offering for a given dataset as well as comparing results across multiple offerings. Using MLflow, BenchML is able to remain cloud-agnostic and offer a delightful local experience while leveraging the aforementioned integration to provide Azure users with a fully managed experience.

Bio: Akshaya is an engineer in the AI Platform at Microsoft, having released both GA versions of Azure Machine Learning over the years and the OSS repo MMLSpark. As the recent version of Azure ML pivoted to become more of an open platform rather than a managed product, his focus has shifted outward for open-source platform definitions for cloud-scale implementations and focused on MLflow for the Azure ML managed tracking store.

Talk 3 - Title: TensorFlow(X) Data Validation: Better ML through better data.
Presenter: Alkis Polyzotis, Google
Abstract: Analysing and validating the input data is a necessary step before building any model. Moreover, any attempt to understand and debug model behaviors inadvertently requires some degree of data understanding. To this end, we have developed Tensorflow Data Validation (TFX) as a library for ML data analysis. This presentation will cover the salient functionality of TFDV, how it is integrated into TFX to support user workflows on model development and debugging, and our ongoing work to bring closer data and model analytics.

Bio: Alkis Polyzotis is a research scientist at Google Research, leading the data-management projects in the TensorFlow Extended (TFX) platform. His interests include data management for machine learning, enterprise data search, and interactive data exploration. Before joining Google, he was a professor at UC Santa Cruz. He has a Ph.D. in Computer Sciences from the University of Wisconsin, Madison and a diploma in engineering from the National Tech. University of Athens, Greece.