The Great MLOps Debate End to End ML Platforms vs Specialized Tools -

Your machine learning team needs at least some basic tooling in order to be effective, but which way do you turn: End-to-end platforms or specialized tools?
--
Over the last few years, it’s been established that your ML team needs at least some basic tooling in order to be effective, providing support for various aspects of the machine learning workflow, from data acquisition and management, to model development and optimization, to model deployment and monitoring.
But how do you get there? Many tools available off the shelf, both commercial and open source, can help.
At the extremes, these tools can fall into one of a couple of buckets. End-to-end platforms that try to provide support for many aspects of the ML lifecycle, and specialized tools that offer deep functionality in a particular domain or area.
At TWIMLcon: AI Platforms 2022, our panelists debated the merits of these approaches in The Great MLOps Debate: End-to-End ML Platforms vs Specialized Tools.
🔔 Subscribe to our channel for more great content just like this: kzbin.info...
🗣️ CHAT WITH US!
===============================
Subscribe to the TWIML AI Podcast: twimlai.com/po...
Join our Slack Community: twimlai.com/co...
Subscribe to our newsletter: twimlai.com/ne...
Want to get in touch? Send us a message: twimlai.com/co...
🔗 LINKS & RESOURCES
===============================
TWIMLcon AI Platforms: 2022 - twimlai.com/co...
Growth Hacking Sports with Machine Learning with Noah Gift - #158 - twimlai.com/go...
The Fallacy of “Ground Truth” with Shayan Mohanty - #576 - twimlai.com/go...
📸 Camera: amzn.to/3TQ3zsg
🎙️Microphone: amzn.to/3DkKSGh
🚦Lights: amzn.to/3TQlX49
🎛️ Audio Interface: amzn.to/3TVFAIq
🎚️ Stream Deck: amzn.to/3zzm7F5
Disclaimer: Some of these links are affiliate links, we’ll make a small commission if you make a purchase at no additional cost to you.

[@chris.dillon]

I don't know why I'm leaving this comment. No idea. Enterprise is just an adjective modifier for crappy software. Microsoft has been trying to undo decades of end-to-end and embrace open things. The only thing I care about with Enterprise software is does it have roles/authz/multi-tenant. Everything else is horrible experiences that get disrupted, always has been. We never got devops in a box. The only box we could have drawn was around the earth. We wanted devops in a box because the whole CNCF foundation eye-chart of options was disturbing, if not easily mistaken for a Pokemon collection list. "We just deployed Butterfree proxied by Metapod". MLOps is just devops applied to ML. Same maturity crisis. The only thing new, novel, mysterious and special is the data. I have no answers or opinions on that piece, I'm out of my league. The data management, scale, generation and the rest is all new ground to me and I have no role models or projects to cargo cult.