Capgemini + MongoDB on agents
A new Capgemini + MongoDB talk frames agentic AI as a layer on top of robust data pipelines, stressing modular APIs for plugging agents into existing workflows and the need to manage data locality and compliance. The discussion highlights the practical challenge of connecting agents to operational data stores at enterprise scale. (youtube.com)
The on‑demand webinar was presented by Steve Jones, Executive VP for Collaborative Data Ecosystems at Capgemini, and Prasad Pillalamarri, Director for SI Partner Solution Consulting at MongoDB, and the demo explicitly used the Model Context Protocol (MCP) to surface agent context to downstream tools. (mongodb.com) MongoDB emphasized a bidirectional agent‑database pattern where agents perform semantic retrieval via Atlas Vector Search and write session context and tool‑call metadata back into documents; the demo and examples included a 1,536‑dimensional embedding query and JSON session documents with messages and tool_calls_made fields. (dev.to) The joint offering frames operational readiness as more than models: Capgemini’s RAISE capability includes governance, real‑time monitoring and orchestration for agent fleets, and the partnership says it is building over 100 bespoke agentic solutions with NVIDIA while offering three implementation routes—off‑the‑shelf, custom, and embedded agents. (capgemini.com) Capgemini Research Institute survey data cited in the conversation projects a potential $450 billion economic opportunity by 2028 for AI agents, but found only 2% of organizations have deployed agents at scale and expects about 15% of business processes to reach semi‑ or full autonomy within 12 months. (capgemini.com) MongoDB’s blog and webinar collateral call out concrete infra gaps: 59% of C‑suite respondents said data is somewhat or completely siloed in a Workday Global survey, and MongoDB positions its document model plus native vector search as a way to remove external vector DBs and reduce RAG latency and pipeline complexity. (mongodb.com) To address developer experience and adoption the ecosystem already offers training and tooling—MongoDB University lists an “AI Agents with MongoDB” skill badge and Capgemini highlights rapid prototyping frameworks and an agent gallery for reusing preconfigured workflows. (learn.mongodb.com)
