Complete your Databricks User Groups profile!

Fill out a few details about yourself so the community can get to know you.
Databricks Manufacturing & High-Tech User Group

What manufacturing data + AI problem are you working on right now?

Summary: Donald Webster welcomes members to the Databricks Manufacturing & High-Tech User Group, focusing on real-world problems and architectures in the field. They invite participants to share their biggest challenges related to data, engineering, analytics, or AI, such as integrating various data systems, managing data across multiple sites, advancing predictive maintenance, and enhancing quality and efficiency. The discussion aims to shape future sessions and technical topics for the group, encouraging participants to share their roles and challenges.
AI Summary

Welcome to the Databricks Manufacturing & High-Tech User Group.

I want this community to be built around real problems, real architectures, and lessons from people doing the work.

To get the conversation started, I’d love to hear what everyone is working on.

What is the biggest data, engineering, analytics, or AI challenge you are currently trying to solve?

A few examples:

• Connecting ERP, MES, historian, maintenance, quality, and supply-chain data
• Managing data across multiple plants or business units
• Moving predictive maintenance from pilot to production
• Improving quality, yield, throughput, or downtime
• Building stronger governance and business context for AI
• Applying Genie, Unity Catalog, Lakeflow, Lakebase, or other Databricks capabilities to industrial use cases
• Preserving operator and engineering knowledge
• Modernizing legacy data environments without ripping everything out

Share your role, industry, and the problem you are working on.

I’ll use the discussion here to help shape upcoming sessions, speakers, and technical deep dives for the group.

Looking forward to learning from everyone.

1 comment
William OSullivan
2 weeks, 5 days ago

I’m working on what I think is one of the harder Industrial AI problems: connecting manufacturing context across ERP, MES, quality, laboratory, historian, maintenance, and engineering systems well enough for AI to reason over it reliably.

The challenge is not just getting the data into one place. It is preserving the operational meaning around the data: asset, product, lot, process step, specification, genealogy, quality status, and who has authority to act.

The use case I’m most interested in right now is closed-loop manufacturing:

Design / Plan → Execute → Measure → Decide → Improve

How do we take evidence from production and quality, connect it back to engineering and planning, and then use governed AI agents to investigate issues, correlate evidence, and recommend actions without bypassing the people and systems that own the decision?

My role is CEO of Connected Manufacturing, working primarily with complex manufacturing environments.

I’d be especially interested in hearing how others are approaching the context layer between the lakehouse and MES/OT systems. That feels like the part of the architecture that will determine whether industrial agents become genuinely useful or remain impressive demos.

Donald Webster
1 month ago