Coffee Sessions #43 with Kyle Gallatin of Etsy, Maturing Machine Learning in Enterprise.

Join the Community: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://go.mlops.community/YTJoinIn⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠

Get the newsletter: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://go.mlops.community/YTNewsletter⁠⁠⁠⁠⁠⁠⁠⁠⁠

// Abstract
The definition of Data Science in production has evolved dramatically in recent years. Despite increasing investments in MLOps, many organizations still struggle to deliver ML quickly and effectively. They often fail to recognize an ML project as a massively cross-functional initiative and confuse deployment with production. Kyle will talk about both the functional and non-functional requirements of production ML and the organizational challenges that can inhibit companies from delivering value with ML.

// Bio
Kyle Gallatin is currently a Software Engineer for Machine Learning Infrastructure at Etsy. He primarily focuses on operationalizing the training, deployment, and management of machine learning models at scale. Prior to Etsy, Kyle delivered ML microservices and led the development of MLOps workflows at the pharmaceutical company Pfizer. In his spare time, Kyle mentors data scientists and writes ML blog posts for Towards Data Science.

--------------- ✌️Connect With Us ✌️ -------------
Join our Slack community: https://go.mlops.community/slack
Follow us on Twitter: @mlopscommunity
Sign up for the next meetup: https://go.mlops.community/register

Connect with Demetrios on LinkedIn: https://www.linkedin.com/in/dpbrinkm/
Connect with Vishnu on LinkedIn: https://www.linkedin.com/in/vrachakonda/
Connect with Kyle on LinkedIn: https://www.linkedin.com/in/kylegallatin/

// Takeaways
Data science is still poorly defined, and there is a large variance in organizational maturity  
Basically, everything we need for mature ML in modern organizations exists technically, except for the strategy, mentality, organization, and governance
Organizations that poorly define data science often overburden their data scientists, but there are expectations that data scientists know some engineering
Operationalizing data science is not that different from software engineering, and software engineering can be one of the most valuable skill sets for a data scientist.

// Q&A with Kyle as a data science mentor:  
https://www.youtube.com/watch?v=7byRQGHD39w&t=1s

Timestamps:

[00:00] Introduction to Kyle Gallatin
[01:00] Kyle’s Path into Tech
[02:04] Data Analyst to Engineer
[03:45] Reflections on Learning CS
[04:04] SAS App with ML Services
[05:13] Python’s Strength in Machine Learning
[06:43] Working Effectively with YAML
[07:10] Choosing Technologies and Plug-ins
[08:43] Take the Easy Way
[09:00] Favorite Plug-ins Overview
[09:07] VS Code Remote SSH
[09:44] Future of Machine Learning
[11:12] MLOps Growth and Buzzword Status
[12:08] Exploring Heuristics and Next Steps
[15:19] Navigating Unknown ML Territory
[15:33] Monitoring and Observability Practices
[16:21] Specialized and Customized Solutions
[17:43] Balancing Commonality and Specificity
[17:54] Integrations Across ML Systems
[20:00] Measuring Time to Production
[20:22] Data Scientists’ Team Fit
[21:34] One Size Doesn’t Fit
[22:40] Building Depends on People
[23:40] Defining Data Science Roles
[24:00] Platform Engineering Perspective
[25:00] Optimizing Model Serving Value
[25:21] Model Serving Platforms
[27:13] Importance of Standardization
[29:00] Exercising Good Judgment
[29:57] Breaking Work into Pieces
[30:30] Data Access Regulations
[33:32] Technical Standpoint Discussion
[34:37] Defining Use Cases Clearly
[36:04] Next Big Thing: MLOps
[37:50] Modern Scaling Approaches
[38:46] Nontechnical Companies Stepping Up
[41:18] Defining Core Problems
[42:38] Considering Value and Needs