DiscoverThe Ravit ShowBuilding AI tools that can work in production
Building AI tools that can work in production

Building AI tools that can work in production

Update: 2025-11-02
Share

Description

I’m at Big Data London and just wrapped a conversation with Jordan Burger, Applied AI Lead at Keboola. He’s focused on building AI that data teams can trust in production, and he’s one of the builders behind Keboola Data Agents.


Here’s what we got into:

* Why a full data platform matters when AI needs real context, not just a few tables

* The lifecycle context agents need that standalone tools never see

* What breaks when AI only sees fragments of the data story

* How platform-native AI handles messy data, lineage, policies, and tribal knowledge

* Why context-aware implementation can matter more than the model itself

* The role of historical data stores for agents and what this means for future architectures


Jordan is on stage tomorrow to showcase this work. Both agents are live on the Keboola platform. Link in the description to spin up your first flow with your own data.


If you’re at Big Data London, meet the Keboola team at Booth L40.


#data #ai #bigdataldn2025 #keboola #theravitshow

Comments 
00:00
00:00
x

0.5x

0.8x

1.0x

1.25x

1.5x

2.0x

3.0x

Sleep Timer

Off

End of Episode

5 Minutes

10 Minutes

15 Minutes

30 Minutes

45 Minutes

60 Minutes

120 Minutes

Building AI tools that can work in production

Building AI tools that can work in production

Ravit Jain