# From Systems of Record to Systems of Context — Omri Bruchim, monday.com

## Executive summary

The talk outlines a fundamental shift in enterprise software from 'Systems of Record'—which merely log what happened—to 'Systems of Context,' which actively understand the meaning and connections between data points. monday.com addresses this challenge by building a 'Monday world model' using two distinct, precomputed engines: a slow engine for durable, long-term user profiling (who you are) and a fast engine for real-time signals (what is urgent today). This architecture mirrors both the brain's hippocampus/neocortex function and the Lambda data architecture, ensuring that AI agents can reason on deep context rather than just retrieving disconnected data.

## Key takeaways

- The Problem: Data vs. Understanding: Current AI assistants have access to vast amounts of data (boards, emails, Slack messages) but often fail because the core problem is not retrieval (data availability), but 'understanding'—the ability to connect disparate pieces of information and determine true user priority.
- The Solution: System of Context: Instead of just logging events, the system must build a context layer ahead of time. This 'Monday world model' allows AI agents (like Sidekick/Psyche) to understand not only *what* happened but *why* it matters and *who* is affected.
- Architectural Design: Two Engines: The context layer is built using two engines running on different time windows. The slow engine mines weeks of activity to build a durable profile of user patterns and goals, while the fast engine processes recent data for live signals and immediate urgency.
- Resilience and Intelligence: This dual-engine approach provides resilience; if one source fails, the system degrades gracefully. Furthermore, it allows the AI to understand when to be proactive (notify) and when to remain silent.

## Technical details

- System Architecture: The system utilizes a dual-engine approach, drawing parallels between the brain (hippocampus/neocortex) and data processing. The slow engine handles long-term pattern mining, while the fast engine processes short, recent windows for real-time signals.
- Data Model Components: The model constructs three key areas of context: 1) How the user's work is structured (key entities and relationships); 2) A current snapshot of live signals (overdue tasks, urgency); and 3) Durable profiles detailing work patterns and decisions over time.
- Technical Analogy: The architecture is explicitly compared to the Lambda data architecture, which merges a fast speed layer (real-time) with a slow batch layer (full history).

## Practical implications

- For developers building enterprise AI tools, the focus must shift from simple RAG (Retrieval-Augmented Generation) to deep context modeling and pattern recognition.
- The architecture suggests that combining historical data analysis with real-time signal processing is crucial for creating truly useful 'intelligent' agents.
- Implementing a durable user profile layer based on multi-source activity logs (Slack, email, calendar) can significantly improve agent utility.

## Topics

System of Context, AI Architecture, Data Modeling, Lambda Architecture, Enterprise AI, Workflow Automation, monday.com, Sidekick/Psyche

Source: https://www.youtube.com/watch?v=Btk8wDUVs74