Spotify's coding agents that saved 10 engineering weeks, how Meta encoded tribal knowledge for AI agents, and Pinterest's CDC-to-Iceberg architecture explained
Reading about the agentic AI journey and the future of Data Engineering in this era is so interesting.
> 82% of engineers use AI tools daily, but 64% of organisations are still experimenting or using AI for tactical tasks only
This, and Meta's project to systematically extract and encode undocumented tribal knowledge, really shows that two topics will dominate conversations in the near future: driving adoption and moving fast enough for large companies that may struggle to keep pace, combined with context engineering.
Smazing selection of topic Chozhan, Sri, & Ananda 💛!
Spotify automating 240 PRs to save 10 engineering weeks is the kind of concrete number that cuts through the 'AI for productivity' noise. The 82% daily usage vs 64% strategic deployment gap stands out more: most teams are using AI as a faster keyboard, and the semantic layer vs text-to-SQL debate you raise is the first sign of organizations treating it as infrastructure. The jump from 5 context files to 59 at Meta is also doing a lot of work — is that effort tracked as engineering overhead or as product investment? At theaifounder.substack.com I've been following how AI-native teams handle this transition from individual tool adoption to organizational capability. What's the forcing function that actually flips teams from the 82% bucket into the 64% strategic one?
Reading about the agentic AI journey and the future of Data Engineering in this era is so interesting.
> 82% of engineers use AI tools daily, but 64% of organisations are still experimenting or using AI for tactical tasks only
This, and Meta's project to systematically extract and encode undocumented tribal knowledge, really shows that two topics will dominate conversations in the near future: driving adoption and moving fast enough for large companies that may struggle to keep pace, combined with context engineering.
Smazing selection of topic Chozhan, Sri, & Ananda 💛!
Thank you so much for the mention, it’s an honour to be part of this list and the Data Engineer Things.
Spotify automating 240 PRs to save 10 engineering weeks is the kind of concrete number that cuts through the 'AI for productivity' noise. The 82% daily usage vs 64% strategic deployment gap stands out more: most teams are using AI as a faster keyboard, and the semantic layer vs text-to-SQL debate you raise is the first sign of organizations treating it as infrastructure. The jump from 5 context files to 59 at Meta is also doing a lot of work — is that effort tracked as engineering overhead or as product investment? At theaifounder.substack.com I've been following how AI-native teams handle this transition from individual tool adoption to organizational capability. What's the forcing function that actually flips teams from the 82% bucket into the 64% strategic one?