AWS Identity Center gives you role-based access control and SSO with low effort, giving you a leg up on the road to SOC 2 compliance

DeepStructure developed an application to address our internal IT oncall needs using our TypeScript-native AI platform. In this post I will show how we combined MCP, Agents, and Slack (MÁS stack) to solve a real-world problem. While this is a specific demonstration, the DeepStructure platform allows external developers to mix-and-match MCP servers, different models, and a variety of frontends to create their own agent workflows.

We describe three parallels between MCP and npm: (1) sure, you _could_ build the tools yourself, but now you don't have to; (2) the integrations become the vendor's problem, not yours; (3) even though the ecosystem isn't uniformly high quality, there's still value for developers who follow best practices

We describe "Agentic" systems as Level 1 (intelligent decision making), Level 2 (creative branch generation), and Level 3 (intelligent dynamic execution)

We adopt a rating system for the technical content of blogs by borrowing from the ski slope rating system. From least to most technical: 🟢, 🟦, 🎓, and 🎓🎓

We show how to build a useful slackbot agent using DeepStructure's Assistants, Slack connector, and TypeScript workflows: key features are drawing on external data sources, using previous questions and answers, and adapting to reactions in order to improve itself over time

DeepStructure's mission is to solve the hardest problems of building resilient AI systems, empowering you to deliver magical features to your users