We respect your right to privacy. You can choose not to allow some types of cookies. Your cookie preferences will apply across our website.
Insights on AI agents, autonomy, and the future of infrastructure
Delegation is broken because AI has no memory. Agents give you autonomy. Memory gives you continuity. Without both, you're just automating with extra steps.
Mem0 and Zep give your AI agents personal memory. But there's a second type — organizational memory — that nobody is building for. Here's the difference and why it matters.
Klarna's AI saved $60M and broke the company. The missing piece isn't better prompts or more context — it's intent infrastructure.
When AI agents share contextual persistent memory, they stop repeating mistakes and start building on each other's work.
We're releasing an open-source plugin that connects Claude Code to gutt. Your AI agents now know who has expertise, what was decided, and why.
AI coding assistants are stateless by design. The gutt Claude Code plugin introduces persistent organizational memory so your AI learns from your entire team.
Insurance brokers exist because of context. But context doesn't scale. Here's the uncomfortable truth about what happens when growth erodes your competitive advantage.
16 stories, 66 story points, 108 files, 15,741 lines of code. One enterprise backoffice portal. 20+ parallel agents. 4 hours. No detailed specs. Just plain English.
In August 2025, I told our team we were dinosaurs. Six months later, I proved we could be raccoons—shipping 90 hours of expert-level work in 3 hours.
Organizations lose their memory as they scale. GUTT is building the context layer that helps AI agents—and humans—work with institutional knowledge.
See how gutt helps teams build AI agents that remember context, learn from past interactions, and never lose institutional knowledge.
Book a Demo