Key Summit Learnings on IDP and Agentic Workflows – and Firemind’s Approach
At the recent AWS Summit London, a clear theme emerged: organisations are moving from passive GenAI tools to agentic workflows – systems that can understand, act, and adapt with limited manual input.
This shift is especially relevant for Intelligent Document Processing (IDP). What used to be a set of simple automation tasks is becoming a more advanced system for handling documents, extracting information, and helping teams make decisions faster and with better control – driven by GenAI agents designed to minimise manual friction while keeping humans in the loop where necessary.
What IDP now looks like
One of the standout talks focused on the evolution of IDP – shifting from traditional extraction to agent-driven document operations. AWS outlined three key pillars:
-
Core IDP – classification, extraction, validation, transformation.
-
Intelligent Search & Retrieval – natural language interfaces for finding relevant information fast.
-
Agent-Driven Document Ops – low-touch workflows powered by GenAI agents, enabling smarter automation with human-in-the-loop design.
MCP: A standard for secure agent workflows
Model Context Protocol (MCP) gives structure to how AI agents operate. It separates control into three parts:
-
Tools – Things the agent can do (like search or create a document).
-
Resources – Data the agent uses, managed by the application.
-
Prompts – Instructions that guide the agent’s behaviour.
MCP is already seeing strong adoption, with 1,000+ server builders, broad usage across enterprise apps, and a growing open-source ecosystem. It’s not just a proposal – it’s a standard that’s being used in real-world systems today.
This matters in industries like finance, where governance, auditability, and control are required. We’re aligning Pulse with MCP so agent workflows are secure, trackable, and ready to scale.
See It in action: learn more in our upcoming webinar
Join us on Wednesday 14 May (2–3PM BST) for a practical look at how financial services teams are moving from GenAI experiments to real business impact.
We’ll show how Bank of Ireland cut legal team workload by 20% using an agentic IDP solution built on Pulse – and how you can apply the same approach in your own organisation.
Get in touch
Want to learn more?
Seen a specific case study or insight and want to learn more? Or thinking about your next project? Drop us a message!