Martech stacks were built for a different era. As AI agents and humans start to work together, rigid systems are getting harder to adapt. This Thursday, Scott Brinker of chiefmartec joins marketing leaders Meagen Eisenberg from Samsara and Kumar Ram from HP to discuss the shift to a more composable approach and what it takes to evolve your stack over the next 3 to 5 years without starting from scratch. Register to hear how teams are rethinking their architecture for the AI age. https://lnkd.in/giTWd9cQ
In the field, the challenge isn’t that martech stacks are outdated — it’s that they weren’t designed for real-time data + AI-driven execution. As agents become part of go-to-market, composability only works if there’s a unified data foundation underneath it. Looking forward to this session, as well. #databricks
Jonathan nails it — martech stacks weren't built for real-time data + AI-driven execution. The composable agent-first architecture is the only path forward. As AI agents execute marketing workflows with access to customer data, pipeline data, and third-party systems, the governance layer becomes critical: Which agent touched which data, did any sensitive info leave the org, and can you audit every action? At Vinkius (vinkius.com) we built an AI Gateway that adds this to Databricks-powered AI workflows: Zero-Trust DLP, SSRF protection, kill-switch, and 30-day audit trails. Data Intelligence + Governance = enterprise-ready AI stack.
Looking forward to this one as demand is growing rapidly.
Samsara is really doing some amazing work, great to know that Databricks is partnering with the company, always wanted to know more about their tech stack. Sanjit Biswas posts of fleet management real time, its ground level impacts and supply chain are super exciting. I do not know whether Databricks realises this, but with Genie, if used properly, you will be changing data solutioning for ever.
The move to composable setups makes sense, but over time the pressure shifts to keeping data connections clean and reliable. If that layer is not stable, even flexible systems can slowly become complex again.
The composable framing is right, but the real test is whether your messaging channels can act autonomously, not just get orchestrated. Most stacks still treat SMS and voice as dumb output layers instead of giving AI agents control of the full conversation.
Kumar Ram quiet interesting - I do not know whether we work for the same HP or not but I can discuss some of your super interesting questions and perspectives.
Looking forward to this session with Meagen and Kumar!
Completely agree. "Without starting from scratch" is the critical piece here. At Systech Solutions, Inc, we see firsthand that Agentic AI is only as good as the foundation it runs on. Moving to a composable approach is the only way to scale these new tools without ripping and replacing your entire infrastructure. Great panel line-up for a highly relevant topic!