INTUEO LABS INC.
Intueo PlatformYour agents, their context and approved tools.Sage 1.5Model research. In development.Intueo VoiceAI reception, calls and the phone-service portal.

Solutions

AI Receptionist

Answer and qualify calls with an AI front desk.

IVR

Smart routing flows for your phone system.

Resources

App Features

Explore the separate Intueo Voice portal.

FAQ

Common questions about Intueo Voice.

Request a Demo

Talk with us about the best setup for your business.

AI & engineering services

Deploy, connect and operate the systems your business needs.

Managed AI Agents

Setup, deployment and ongoing operation with our team.

Cloud & Web Infrastructure

High-performance digital foundations for the AI era.

Cognitive Growth & Analytics

Data science and predictive analytics for market dominance.

Spatial Computing & AR

Next-generation immersive brand experiences.

Schedule a Consultation

Discuss your enterprise infrastructure needs.

Research & engineeringNotes on our models and agent systems.Insights & newsOur full collection of AI articles and analysis.
About IntueoA Canadian AI technology company.How we workScoping, delivery and ongoing support.ContactCareers
Request access

Intueo / Research & engineering

Memory, tools and verifiable work.

Our work with stateful agents starts with a practical question: what must a system retain, authorize and verify to resume a task without starting over?

Read our field notes

Design decisions

Three distinctions that guide our engineering. These are system problems, not promises about model performance.

01

Persistent context is not a longer transcript.

A conversation records what was said, including abandoned plans and facts that have since changed. Loading more of it does not automatically give an agent an accurate working brief.

The decision
Keep durable context distinct from conversation history: current decisions, useful conventions and unresolved work. Retrieve supporting history when needed. This is external state supplied to a model, not a change to the model’s weights.
The consequence
Selection makes the next session more focused, but also creates maintenance work. A remembered fact can become stale. Corrections, retention choices and the ability to inspect what was kept matter as much as recall.

02

Finding a tool is not permission to use it.

A tool can appear in a catalogue without a working connection. A working connection can exist without an individual agent being authorized to use it. Collapsing these states makes capability look broader than it is.

The decision
Treat discovery, connection and agent authorization as separate questions. Discovery describes what exists; connection describes whether the integration is configured; authorization defines which agent may act within the permitted scope.
The consequence
This adds setup steps, but makes failures easier to explain. A missing connection calls for configuration, not another model prompt. A permission boundary should remain a boundary, not become an invitation to retry around it.

03

An active interface is not evidence of completion.

Text can stream while a task is still waiting on a tool. An interface can disconnect while work continues elsewhere. Neither a moving indicator nor a quiet screen tells us whether the requested change actually happened.

The decision
Separate the state of the conversation from the state of the work. Carry the task, its last known result and its unresolved steps across handoffs. Look for a returned result or a check of the resulting state before describing an action as complete.
The consequence
Recovery should start by checking what happened, not blindly repeating the action. That distinction matters whenever repeating work could create a duplicate record, message or change. Continuity includes knowing what remains unverified.

Field notes

Two first-party accounts of the work behind these questions. Each records its context at publication; historical names and configurations do not define today’s offering.

June 28, 2026

Working with Letta Code across the company

Our account of persistent agent context, reusable skills and continuity between coding sessions. It explains why we treat memory as part of the working system rather than a transcript attached to a model.

June 23, 2026

Building memory into our AI receptionist

Repeat calls exposed a practical problem: a natural voice does not provide continuity. This note describes selecting durable facts, updating them, retrieving context quickly and keeping callers’ information separate. The system was in private testing at publication.

Model research

Sage 1.5

Our model research explores post-training and evaluation for agent workflows: following instructions, using sources and asking for clarification when a request is incomplete.

In development. Release and access details remain subject to evaluation.

Explore Sage 1.5 research

For the current product scope and access information, visit Intueo Platform.

Intueo Platform
Intueo Labs Inc.

A Canadian AI company developing stateful agents and business software, with managed deployment and engineering services.

Products & services

  • Intueo Platform
  • Sage 1.5 · Research
  • Intueo Voice
  • Managed AI Agents
  • Engineering services

Company

  • About Intueo
  • How we work
  • Research & engineering
  • Blog
  • Careers
  • Contact

Stay Updated

Get the latest news and insights from Intueo Labs.

@intueo.ai+1 (604) 256-8253
© 2026 Intueo Labs Inc. All rights reserved.
PrivacyTerms of useAccessibilitySitemap