- Home
- What we build
- AI & Intelligence
- Agentic operating systems
Agentic operating systems
AI that moves work forwardacross your business systems.
AI automation and AI agent development for the systems you already run. We connect your tools and approved knowledge so routine tasks can be completed, exceptions can be escalated, and your team can approve the actions that matter.
Today an inquiry that arrives at night waits in an inbox until someone retypes it in the morning. With an agent, it is recorded and drafted before the office opens.
- Read the inquiryWeb form, after hours
- Checked it against your price list
- Recorded it in your CRMSource and owner set
- Drafted the quote
- Waiting for a person. Sending a quote needs approval.
- Update a recordAllowed
- Send to a customerApproval
- Change a priceNever
Illustration, sample data
Four processes that slow teams down, redesigned.
Work waits for a person to notice it, retype it, check it, or chase it. Here is where that happens, and what changes.
From lead to follow-up.
- Lead arrivesForm
- QualifiedAgent
- CRM updatedCRM
- Quote preparedDraft
- Quote approvedSales lead
- Follow-up scheduledCalendar
The quote waits for approval before it is sent.
What the run leaves behind
Sample- Qualified against your rules
- Priced from the approved list
- Sent only after approval
- Follow-up logged to the record
From request to purchase.
- RequestForm
- Supplier checkedAgent
- Documents validatedAgent
- ERP entryERP
- ApprovalBuyer
A person approves the purchase.
What the run leaves behind
Sample- Supplier record checked
- Documents matched to the request
- Entry drafted in your ERP
- Released only by the buyer
From request to resolution.
- RequestEmail
- Account looked upCRM
- Knowledge retrievedDocuments
- Action takenAgent
- Escalated when unsureSupport lead
Anything it cannot answer from your documents goes to a person.
What the run leaves behind
Sample- Account found before replying
- Answer cited to its source
- Action logged to the record
- Unsure cases handed over with context
From data to decision.
- Data collectedYour systems
- Anomalies detectedAgent
- Report writtenDraft
- RecommendationAgent
- Human decisionLeadership
The report arrives before the meeting. The decision stays at the table.
What the run leaves behind
Sample- Figures read from your systems
- Anomalies flagged with the reason
- Report drafted in your format
- The decision recorded by a person
An assistant answers the question.An agent finishes the job.
Every action is limited by the permissions, integrations, and approval rules you configure. Anything outside them waits for a person, and the agent is not assumed to be right: its work is checked where it matters.
A chatbot answers questions. An agent completes tasks. It books the meeting, updates the record, sends the confirmation, and tells you it is done.
The work runs on people remembering.
- An inquiry arrives.inbox
- Someone notices it.next morning
- Someone types it into the CRM.by hand
- Someone checks the price list.spreadsheet
- Someone writes the reply.waiting
- Someone remembers to chase it.memory
Six handoffs. No record of who did what.
The work runs on the system.
- The inquiry is read overnight.Agent
- It is checked against your documents.Price list
- It is recorded in your CRM.CRM
- A reply is drafted in your format.Draft
- One person releases the queue.Morning
- It is sent, and followed up.Sent
One approval. Every step logged to the record.
The approval queue, the log, and ten steps
- QuoteWaiting
Quote draft, Inquiry #2481
Prepared by Sales agent
- Reply6 min
Reply to a delivery complaint
Prepared by Support agent
- Publish14 min
Product page update, 3 items
Prepared by Content agent
- Pricing31 min
Volume request above the standard band
Prepared by Sales agent
Quote draft, Inquiry #2481
Distributor, Ontario. 14 line items.
Why this needs a personThe requested delivery date is outside the standard window, so the agent held the quote instead of sending it.
- To
- Purchasing manager, Distributor, Ontario
- Subject
- Your quote request #2481
Thank you for the request. The attached quote covers all 14 items on your list, priced under your current agreement.
We can deliver on the date you asked for, once our warehouse team confirms the last item.
- Customer record matched
- Prices read from the approved list
- Stock confirmed for 13 of 14 items
- Delivery date outside the standard window
Illustration, sample data
- Read an inquiry from a returning customer and matched it to their account.Grounded in the account record and the last three orders
- Created the record in your CRM with the source, the owner, and the time it arrived.
- Held. Drafted a reply quoting list price for two of three items, and stopped.The third item is priced by exception, which is on your side of the line
- Released by your operations lead. Sent.
- No reply yet. Followed up once.Rule: twice, and then it stops
- Handed to a person without answering a second inquiry.Nothing in your documents answers it
- Entries
- 6
- Held
- 1
- Handed over
- 1
Illustration, sample data
Scroll the drawing sideways to read it.
One inquiry, ten steps, one of them yours.
Where the line is drawn.
What the system may do automatically
System actsLogged
- Read and sort what arrives
- Check answers against your own documents
- Create and update records
- Draft replies, quotes, and reports
- Follow up twice, then stop
- Log everything it did, and why
What requires approval
System preparesPerson approves
- Anything sent to a customer in your name
- Quotes and proposals
- Pricing outside the published list
- Refunds above a limit you set
- Anything published
What stays human
System briefsPerson decides
- Price and credit decisions
- Contracts and legal commitments
- Exceptions and unusual cases
- Relationships
- The limits themselves
- The off switch
The boundary is agreed in writing before the build, and every action is visible in the audit trail afterwards. The off switch is yours, and it works without calling us.
What AI agent development includes.
The workflow map
The process drawn end to end, with the people who run it.
The written boundary
What the system takes over and what stays with your people, signed before the build.
An agent brief per agent
Role, tools, limits, and what it may never do.
The approval queue
The screen your team works in.
The audit trail
Every action and its reason, in plain language.
The controls
Limits you can tighten yourself and an off switch, both tested before launch.
The data handling document
Written for your privacy officer.
The smallest version: three modules
The smallest version of this
Three fixed-scope modules, sold inside a build.
A full operating system is not the only way in. Three modules have a fixed scope, which means what each one does is fixed in the signed scope of the build. They are modules of a build, not a subscription.
- Follow-UpFixed scope
Quotes and inquiries chased on a schedule you set, twice, and then it stops. Every touch logged against the record.
- Front DeskFixed scope
Inbound read, understood, recorded and routed to the right person, in the hours you choose, with everything sensitive held for approval.
- Site AnswerIncluded with every build
Your site answering a real question from your own documents, with the source shown.
How we assess feasibility and scope.
The agent works across your existing systems through the access you grant it. Every connection is checked by name during scoping, before you sign.
Agent layer
Acts within its permissions- CRM
- ERP
- Accounting
- Calendar
- Documents
- Databases
- Your industry system
The systems your business already runs
When required, a person decides.
- Quotes and pricing exceptions
- Sensitive replies
- Anything published
- Anything that cannot be undone
What the system doesWhat your business already runsWhat stays with a person
Where it pays for itself
- Work arrives outside office hours and waits for the office to open.
- Two systems are bridged by a person and a spreadsheet.
- Volume is seasonal, and the peak is covered with overtime.
- Something is written down: a price list, a policy, documents an answer can be grounded in.
Where something smaller fits
- The process runs a few times a week and one person holds it comfortably.
- The work is one rule with no judgement in it. That is automation, and it costs less.
- Nothing is written down yet. The first project is writing it down.
- The requirement is that nobody reviews anything. Every system we build has a person in it.
Where the answer is the smaller thing, we will say so, and build the smaller thing.
The systems it connects to
CRM
Salesforce, HubSpot, Zoho, DynamicsAccounting and ERP
QuickBooks, Sage, NetSuite, SAPEmail, calendar, and messaging
Microsoft 365, Google Workspace, WhatsApp BusinessBooking
Calendly, Acuity, an in-house dispatch boardPhone
RingCentral, Twilio, your existing phone systemE-commerce
Shopify, WooCommerce, BigCommerceDocuments
SharePoint, Google Drive, DropboxYour industry system
dispatch, estimating, practice management, property managementYour own database
read-only views, scheduled exports, a purpose-built interface
Automation, an assistant, or an agent
Three different things are sold under one word. They cost different amounts, they fail differently, and two of the three are often the right answer.
| Deciding factor | A rule-based automationConfigured | An assistant in a windowLicensed | An agent that finishes the jobBuilt |
|---|---|---|---|
| What it produces | A step that fires when a condition is met. | An answer, for whoever asked. | Completed work, and a record of how it got completed. |
| Who works out the steps | You do, in advance, for every case you thought of. | The person reading the answer. | The system, inside limits you wrote down. |
| A case nobody predicted | Nothing fires, and often nobody notices. | It answers anyway, and sounds just as sure. | It stops, hands over, and the handover is in the log. |
| Where the work ends up | In the tool the rule lives in. | On the screen of whoever asked. | In the systems you already run. |
| What it knows | Whatever is in the trigger. | What it was trained on, plus whatever was pasted in. | Your documents, your price list and your history, with the source shown. |
| Cost to change it later | Low. One rule at a time. | Nothing to change. | Higher, because a change is designed, tested and logged. |
| Cost to start | Lowest. Usually configured rather than made. | Low. Usually a licence you may already hold. | Highest. It is a build. |
| When it is the right answer | The process is one rule, and the rule is stable. | People need to ask questions and act on the answers themselves. | The work has judgement in it, crosses systems, and currently waits for a person. |
Which one to choose
If the process is one rule and the rule is stable, buy the automation. It costs less than anything we would build, it will not drift, and a rule that has run for three years without an exception does not need a system that reasons about it.
If what your people mainly need is to ask questions, an assistant is enough, and most of the software you already own now includes one. Paying for a build to get that is paying twice.
Agents earn their cost in one place, and it is a narrow one: work that has judgement in it, crosses more than one system, and currently waits in a queue until somebody gets to it. The discovery call is where we find out whether your process is in it.
Questions people ask.
-
What happens when it gets something wrong?
The design assumes it will. Sensitive actions wait for a person, anything the system cannot ground in your documents is handed to a person rather than guessed at, and every action is logged so a mistake can be found. The first workflow runs alongside your team, who see and approve everything, so the wrong answers happen where they cost nothing.
-
Are you replacing our staff?
No, and we will not scope a project on that basis. The system takes over the reading, routing, chasing, drafting, re-keying, and reporting. Price, credit, judgement, and relationships stay with your people.
-
Do we have to change our software?
No. The system works inside what you already run. If a tool cannot be connected, you hear it during discovery and the scope is drawn around it.
-
Is our data used to train a model?
No, and that commitment is in the contract. The storage region and the model region are named in the scope, and each processor and what it sees is documented for your privacy officer.
-
What does it cost?
The price is set after the discovery call and written into the proposal you sign, because an honest number needs the scope. Scope, price, and timing are agreed before implementation.
-
What if we want it to stop?
You stop it. The off switch works without calling us, and it is tested before launch. The system, its configuration, and its data stay in your accounts either way.
What a build has to be designed around
A system that reads customer messages and acts on them touches more law than a website does. These are the requirements this service is designed around, and the line between what we build and what your counsel decides.
-
PrivacyPIPEDA, and Quebec's Law 25 where it applies
What we design forA stated purpose for every system an agent can read, the smallest tool list that does the job, consent captured where the agent initiates contact, and retention set per record type rather than left at a default.
What we document for your counselWhat the system holds, where it is stored, who can reach it, how long it is kept, and every place a decision about a person is made with automated assistance.
-
Health informationWhere the work touches it at all
What we design forNo health information in an agent's context until the storage region and the handling are settled. Where they are not settled, the workflow is scoped to exclude it rather than scoped around it.
What we document for your counselWhich data classes were excluded and why, so the boundary is something your privacy officer reads rather than infers.
-
Electronic messagesCASL, for anything automated
What we design forConsent and its basis recorded against the record before any automated follow-up, an unsubscribe that works in one action, and a follow-up sequence that stops rather than loops.
What we document for your counselThe consent basis per contact and the full send log, which is the evidence the legislation asks you to be able to produce.
-
AccessibilityWCAG 2.2 AA, plus AODA or the Accessible Canada Act where they apply
What we design forEvery surface a person actually uses, which on this service means the approval queue, the audit trail and any customer-facing reply. Keyboard, screen reader, contrast and reduced motion, tested rather than asserted.
What we document for your counselThe test results per surface and the standard tested against, so a barrier report can be answered with a measurement.
-
LanguageQuebec's Charter of the French Language, where the business operates there
What we design forPer-language operation as a first design decision rather than a translation pass at the end. Customer-facing text is authored per language, including the automated replies, the error states and the follow-up.
What we document for your counselWhich surfaces are authored per language and which are not yet, so the gap is a list somebody owns rather than a surprise.
-
Where the data livesAnd which processor sees it
What we design forThe storage region and the model region named in the scope before the build, chosen rather than inherited from a default.
What we document for your counselEach processor, what it sees, the region it sees it in, and what changes if you move region later.
Where the business also operates in the United States, the equivalent requirements are named in the scope.
HUREAL designs to a standard, tests against it, and documents what was built. The determination belongs to your counsel or your privacy officer. Our own accessibility target and how to report a barrier are on the accessibility page.
Two ways to start from here.
One process creates too much manual work.
Workflow Opportunity Sprint
We map it, then define what a connected system should do next.
Explore the SprintI know what I need to build.
Discovery call
A free conversation, then a written proposal with scope, price, and timing.
Request a discovery call