Challenge
Sloy Case: Company Working Memory for AI Agents How KT.Team turned chats, meetings, documents, tasks, Git, and finance into machine-readable context for AI agents. 2026-05-07 Discuss the project Key takeaways How KT.Team turned chats, meetings, documents, tasks, Git, and finance into machine-readable context for AI agents. In service and project-based companies, work context usually stays in chats, meetings, emails, Google Drive, tasks, Git, and financial spreadsheets.
The context is stored in a corporate Git repository: README files, monthly summaries, decisions, risks, agreements, and links to source materials. As a result, the agent can answer questions about project status, commitments to the client, payments, margin, risks, and changes after vacation without manually gathering context from a manager. The challenge In service and project-based companies, work context usually remains in chats, meetings, emails, Google Drive, tasks, Git, and financial spreadsheets.
Solution
When an AI agent joins the work, it only sees the current conversation and asks again for the project, client, decisions, and history. Sloy solves this as a corporate memory layer: it collects the work trail from source systems, links it to projects, clients, employees, and money, and turns heavy documents and transcripts into short machine-readable representations.
Corporate memory for AI agents Solution The context is stored in a corporate Git repository: README files, monthly summaries, decisions, risks, agreements, and links to source materials. The agent reads short project memory and returns to the originals only to verify a disputed fragment. Sloy already includes routing scenarios for incoming emails, files, meetings, and Plaud recordings: the LLM suggests a project based on participants and text, a human confirms when confidence is low, and similar sources are then routed automatically.
Review a similar project with an architect
What changed
What changed Result - the agent can answer questions about project status, commitments to the client, payments, margin, risks, and changes after vacation without manually gathering context from a manager.
Blog Articles on the Topic All Articles 10.7.2026 Why Enterprise AI Agents Stall: Production Workflow vs Demo 8.7.2026 Developer and PM Evaluation: L10-L60, Workflow, and AI 28.6.2026 What McKinsey, Gartner, BCG, and Sequoia Say About AI Adoption, and What It Means for You 26.2.2026 How to Build an AI Assistant for Business: Development, Integration, MVP, and Scaling to an Enterprise System 11.12.2025 DORA 2025, Part 7.
Results
How value stream management (VSM) helps IT teams achieve sustainable growth and use AI strategically Projects Related cases All cases OSNO-VA: An AI accountant that handles accounting operations according to regulations Built an AI platform #1c #ai #bi #consulting #cost 2026-06-18 Learn more Financial agent on MCP, SQL, and executable skills #ai #bi #consulting #integration #superset 2026-06-04 Learn more Case tags Click on
tag opens the case catalog filtered by this tag #ai #consulting #integration All cases
