Making Agentic AI Enterprise-Ready: The Memory and Context Challenge

​Agentic AI is moving fast from prototypes to real enterprise use, but one major challenge remains: memory, context, and how agents actually learn and operate over time instead of just responding to prompts. At the Applied Innovation Exchange, this event will bring together startup founders, builders, and enterprise leaders to explore what it really takes to make Agentic AI enterprise-ready.

​The discussion will cover both business and technology perspectives, from new enterprise architectures and agent platforms to orchestration frameworks, memory layers, and multi-agent systems.

​If you are working with or interested in ecosystems around agent frameworks, retrieval and memory systems, and the next generation of AI application stacks, this will be an evening to meet the people building this space and to discuss where enterprise AI is actually heading next.

​In partnership with AICamp: https://www.aicamp.ai/event/events

​Speakers:

​Jerry Liu, Co-Founder and CEO, LlamaIndex

​Jerry is the co-founder/CEO of LlamaIndex, a company that is building agentic document processing. Before this, he led the ML monitoring team at Robust Intelligence, did self-driving AI research at Uber ATG and worked on recommendation systems at Quora.

​LlamaIndex is a fast-growing, Greylock/Norwest-backed company that provides best-in-class document OCR and workflows, powered by AI agents. We provide technology that can read and extract context from the most complicated documents - PDFs, Powerpoints, Word, Excel - letting you convert unstructured context into agentic knowledge work automation.

​João (Joe) Moura, Founder and CEO, CrewAI

​João (Joe) Moura is CEO & Founder of CrewAI, a multi‑agent orchestration framework and platform. He has built in roles previously at Clearbit and others, and is deeply immersed in how teams, developers, and organizations are starting to delegate decision‑making and task execution to groups of agents. Joe focuses on practical orchestration: what it means for enterprises when tools don’t just assist, but act, automate, coordinate, and integrate across workflows.

​CrewAI is a multi‑agent orchestration platform/framework, enabling the coordination of autonomous AI agents to perform complex tasks or workflows. It’s used by a considerable number of large enterprises and supports scaling up agent‑based automation. For Capgemini, CrewAI represents significant opportunity: automating business workflows, improving client delivery, enabling novel services, or even partnering to build agent orchestration when needed.

​Tara Khani, Co-Founder and CEO, Moorcheh.ai

​Tara Khani is Co-Founder and CEO of Moorcheh.ai, a company building enterprise-grade knowledge and memory infrastructure for AI systems. With a background spanning enterprise technology, cloud platforms, and large-scale customer environments, Tara focuses on how organizations can operationalize AI securely and at scale—bringing together data, infrastructure, and governance into production-ready systems.

​At Moorcheh, Tara leads the vision of deploying fully integrated, sovereign AI infrastructure directly within enterprise environments. The platform enables organizations to ingest, manage, and retrieve large volumes of multimodal data with high performance and strict security guarantees. For Capgemini, this represents a critical layer in enterprise AI architecture: scalable, secure memory and retrieval systems that underpin reliable agentic applications and data-driven workflows.

​Majid Fekri, Co-Founder and CTO, Moorcheh.ai

​Majid Fekri is Co-Founder and CTO of Moorcheh.ai, where he leads the development of next-generation AI retrieval and memory infrastructure. With over 15 years of experience building large-scale data systems across companies such as The Weather Network, Prodigy, Rogers, and SOTI, Majid brings deep expertise in turning complex data into production-ready systems.

​At Moorcheh, Majid focuses on solving one of the core challenges in enterprise AI: reliable, scalable retrieval. His work is grounded in information-theoretic approaches to indexing and search, enabling deterministic, high-precision retrieval at scale. This directly addresses issues such as context degradation and inconsistency in traditional RAG systems. For Capgemini, Moorcheh’s approach represents a foundational capability—ensuring that AI systems and agents operate on accurate, high-quality context, even under enterprise-scale workloads.