Skip to content
AtheronLABS

You're visiting from the United States. Prices are shown in US dollars. Not right?

labs@atheron:~/locations/ottawa/ai$ train agent --grounded

AI for Ottawa teams

Assistants that answer in both languages.

Public and member-facing organisations hold policies, guidance and forms that people struggle to find. We build assistants that answer from those documents in English and French, cite them, and hand over to a person when they should.

// why it differs here

What matters for this work here.

01

Answers in the reader's language

An assistant answers in English or French, from the same approved documents, and cites the page in the language the reader asked in.

02

Kept in Canada

We can run an open-source model on servers in Canada, so questions and documents never leave the country or reach a third-party model provider.

03

Limits you can read

Every agent we build has written limits: what it may do, what it must ask a person about, and a log of what it did. It always says it is an AI.

// example project

One we could build for you.

Hypothetical, labelled as such, and priced from our rate card when you open the page.

Example projectGovernment and public sector

Plain-language resident assistant

An assistant on the public website that answers from published bylaws, fee schedules and service pages in plain language and links to the source, on an open-source model hosted in Canada in an account the government controls, so no resident's question leaves its jurisdiction.

// Example project

Build, In Canadian dollars
CAD 159,300 to 243,500, delivered within 23 weeks

// questions

What buyers here ask us.

Will an assistant make things up?

We ground it in your own documents and test it against real questions before launch. When it does not know, it says so and hands over to a person.

How do you set an agent's limits?

In writing, before we build: what it may read, what it may change, what needs a person's approval. Our guide on agent guardrails goes through it.

What does it cost to run?

The estimate splits the one-off build from the monthly running cost, so you see both before you commit.

// next step

Tell us what you need.

A short description is enough. We reply with questions, a range and a time to talk.

AI development for Ottawa: assistants, agents, models | Atheron Network Labs