Would love to hear more about this if any more info comes to light.
I’m sure they’ll have more infor as its finalized but I believe it is still in the early stages of architecture at least that’s what I understood from the presentation.
@josecgomez I think that was one of our EpiUsers podcast episodes too.
Yes, i have marked this as delivered. Thanks.
agree… a two week implementation is not desired. it takes longer than that to train anyone on how to log in and find the first screen (just kidding)… but 90 days is not out of the question. my very first implementaiton was 90 days (plus the 8 weeks we took for education at the beginning). we started Jan 31, with education… got the software installed April 1, and were done with the Pilot on May 31… we had the entire month of june to finish the data migration, and went live july 1 with a FULLY TRAINED team.
We have a very high profile new customer implementation that is very large and very critical and very complex (with very few customizations). They are scheduled for a no-failure optional implementaiton at the beginning of September. so… it can be done. (
Safe Harbor )
And there it is… Get the right mindset and customer team, I could well imagine it is.
The Cloud SDK session that @josecgomez presented was helpful, even though it mostly confirmed what I already encountered.
We have a bunch of custom services created with the legacy SDK that will have to be ported to the cloud-friendly version, and I’ve been puzzling over the best way to do this.
The biggest problem I see with the Cloud SDK services/BPMs/Functions solution is lack of support for development tools, primarily source control.
We have all of our custom SDK code in a git repository with a bunch of Visual Studio solutions. Losing the ability to track source code changes will easily make it unmanageable.
Does anyone have a link to a GROW demo site from insights with login?
Another data point for @pferrington and @tguille_epicor who have been gathering requirements for DevOps items.
See below for clips from my presentation. https://www.youtube.com/watch?v=NTF16Et6iTg Thanks @Randy for taking the video. Full recording coming soon. Don’t ask me why AI selected these clips.
During one of the BPM sessions I attended, one thing that came up was how to manage security secrets in Kinetic (credentials, API keys, etc). We have a number of customizations that make API calls to external systems, and there isn’t a good way to store required secrets for those out of the box (that I know of).
We did a joint presentation this year with Epicor and there were some technical glitches. Check out our BISCIT ERP Solutions - YouTube page for some of our capabilities.
If you have more questions, use this link to put 30 minutes on my calendar and I will answer any questions that didn’t get covered in the presentation.
@Mark_Wonsil has an idea for improving this, I think. It’s been a limitation forever.
Thank you everyone for this “serious” thread. This has some great nuggets…
Note that every year, we also have 3rd party media attend. This year was no exception except, that we had a record breaking year for the number of media and analysts attending the event. One has already written and published their article about the Keynote… it can be found here. For anyone who didnt attend, this really highlights much of what was said: Epicor takes a community approach to Agentic AI -
That’s true. It’s an Epicor Idea.
Implement Secrets Management - KNTC-2555
The challenge is how to you give access to an Epicor Function that runs under one credential (that I’m aware of) and grant secret access is a fine-grain way. 
Hey Tim,
I think I have another Idea that Epicor has delivered.
Seems a good place to hype these ideas too.
TWO WEEKS!!! Cause…well… 
Thanks Mark… the reason it was not marked as delivered is because the idea was never “attached” to the epic that made it happen. I try to catch them all, but this one slipped through.
I had a limited experience at Insights this year. I had some courses on Sunday and monday, and made it to the opening spread and walked around a bit. I also made it to birds of a feather to see some of the awards etc and listen to some of the speakers… then I was struck down with the flu and was absolutely sideways for all of 8 days (it was brutal).
the one thing I will say is that I did attend a prism breakout group to listen to the functionality… and it seemed to me at least from what I got out of this breakout group that all prism seems to do is use my own baq’s to query data for me… and that’s all something that has to be setup… which is a bit of a let down. Why do we need to pay for prism if its using my own baq’s? Why do I need prism to look at my baq’s when im perfectly capable of making them and utilizing them?
I’m much more interested in using an API so that I can access Epicor data headlessly via telegram or some other chatbot. If I can use use a harness to pull data for me it would make analyzing that data quicker and easier in my opinion… Prism seems to be far off from providing the sort of value I would expect… but there I go expecting again…
Gomez has a good talk at birds of a feather, and I apologize but I forget the other guys name - they were both really good - but im foggy on the details now after being down and out for a week with the flu in may…