/ 01FAQ
Frequently asked questions
Practical answers about how we work, what we deliver, and how to get started. If you don’t find what you need, reach out and we’ll reply quickly.
/ 02 Engagement
How we start
What kind of work do you take on?
Integration — wiring off-the-shelf agents (Codex, Claude Code, Hermes, Cowork, OpenClaw) into your real systems. We don’t build a platform; we make the frontier useful in your tools.
How do we start?
Find your row on the readiness map on the services page. We reply within 48 hours with the format that fits and a scoped first step.
What engagement formats do you offer?
Four: Advisory (half-day), Sprint (2 weeks, fixed scope), Build (4–12 weeks, milestones), and Operate (monthly retainer — pause or end any month).
/ 03 Delivery
What you receive
What do you deliver at the end of a mission?
Something that runs without us: wired-in agents, skills, eval suites, runbooks, MCPs, hooks, and memory stores. All yours from day one.
Which agents and models do you work with?
Off-the-shelf agents on top of Anthropic, OpenAI, Google, and open-weights models. We pick the best one today and re-pick on a cadence as the frontier moves.
Do you have a proprietary platform?
No — and that’s the point. The best model has a shelf-life of weeks; the skills, evals, and tools we build are portable and yours to keep.
/ 04 Operations
In practice
Can you work with sensitive data?
Yes. Your skills, keys, and data stay yours. We use least-privilege access, keep environments separated, and operate under NDA when needed.
Who do we call when it breaks?
Whoever you’d call now. We hand over code, skills, ops, and an on-call book to your team — retainer is optional, never required.
Where are you based?
Montréal, remote-first — we make it work across time zones wherever you are, and travel on-site when it matters. We work in English and French.
/ 05 — Next step
Still have questions?
Drop us a line — we'll get back to you quickly.