Context Engineering 2.0: MCP, Agentic RAG & Memory // Simba Khadder

2026-02-28 · Source: MLOps.community · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Software Development & Engineering · Depth: Intermediate, extended

Summary

Simba, from Reddis, introduces "Context Engineering 2.0," a concept focused on unifying disparate context types for AI agents and Large Language Models (LLMs). Currently, Retrieval Augmented Generation (RAG) for unstructured data, memory management (short-term and long-term), and structured data access are treated as separate concerns, leading to "toolbloat." Reddis, known for real-time data delivery, is now focusing on real-time context for agents. The presentation critiques common approaches for structured data, such as text-to-SQL (citing security, performance, and accuracy issues) and API wrapping (due to complexity and token inefficiency). Instead, it proposes a schema-first, business-layer abstraction approach, defining entities, attributes, and relationships to create a navigable semantic catalog. This unified approach, leveraging Reddis's existing capabilities in RAG, memory, and structured data retrieval, aims to provide a single pane of glass for all context, enabling agents to access and reason about data more effectively.

Key takeaway

For CTOs and VPs of Engineering building AI agentic workflows, your teams should prioritize unifying context sources like RAG, memory, and structured data. Adopting a schema-first approach to define data models for agents, rather than relying on risky text-to-SQL or inefficient API wrapping, will enhance security, improve observability, and enable more robust agent performance by providing a single, coherent view of all necessary context.

Key insights

Unifying RAG, memory, and structured data into a single "context engine" is crucial for effective AI agents.

Principles

• Context is paramount for agent intelligence.
• Schema-first design simplifies agent data access.
• Materialized views enhance security and observability.

Method

Define a business-layer schema for entities, attributes, and relationships, creating a navigable semantic catalog that serves as a unified context engine for agents, rather than relying on text-to-SQL or API wrapping.

In practice

• Avoid direct agent access to production databases.
• Implement a unified semantic and access layer.
• Utilize open telemetry for context observability.

Topics

• Context Engineering
• AI Agent Context
• Structured Data Access
• Retrieval-Augmented Generation
• Redis Context Engine

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Engineer, Machine Learning Engineer, AI Architect

Related on AIssential

• Context architecture is replacing RAG as agentic AI pushes enterprise retrieval to its limits — AI agents require real-time, machine-optimized context and memory, moving beyond human-scale retrieval systems.
• I Built an AI Data Engineering Mentor from Scratch: Here’s What Nobody Tells You — AI agent effectiveness is primarily driven by structured context, not just the underlying large language model.
• Mergeable by default: Building the context engine to save time and tokens — Peter Werry, Unblocked — Context engines optimize AI agent performance by providing highly relevant, curated organizational context, moving beyond simple data access.
• Unlocking the Data Layer for Agentic AI with Simba Khadder — Dynamic context engines, built on materialized views and semantic layers, are crucial for scaling autonomous AI agents.
• How RAG, GraphRAG, and Context Engineering Improve AI Performance — Effective AI performance hinges on context engineering, not just raw model intelligence.
• What Is Agentic Storage? Solving AI’s Limits with LLMs & MCP — Agentic storage provides persistent, purpose-built memory for AI agents, overcoming LLM statelessness and RAG's read-only limitation.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by MLOps.community.