The questions everyone asks.
Honest answers to the things every prospect asks us. If yours isn't here, drop us a note.
How long does an AI project take?
Most pilots ship in 3–6 weeks. Full deployments run 6–14 weeks depending on scope. We can usually have a working prototype in 2 weeks.
Who owns the IP we build together?
You do. We sign full IP assignment on every engagement. Your models, your data, your code.
Do you train shared models on our data?
Never. Your data is never used to train shared AI models and is kept logically isolated per tenant.
What size company do you work with?
Anywhere from a 3-person shop to a publicly traded enterprise. About half our clients are SMBs (under 250 employees).
What's the smallest project you'll take?
Our $4,800 Starter Sprint is the floor. It's perfect for small businesses validating their first AI use case.
What's your security posture?
Security-hardened by design: data encrypted in transit and at rest, logically isolated per-tenant data, scoped access controls, signed webhooks and OAuth state, and rate limiting. We follow GDPR-minded data practices and never train shared models on your data. Formal certifications such as SOC 2 are on our roadmap, not yet held.
Where is our data hosted?
The AI-Team platform runs on our managed cloud. Region choice, GCP/Azure, and on-prem or air-gapped deployments are available for custom enterprise engagements — contact us and we'll be straight about what we can support for your requirements.
What models do you use?
AI-Team is powered by Anthropic's Claude. For custom engagements we can architect for model portability across providers (OpenAI, Google, Meta, Mistral, and others) so you're never locked in.
Will we be locked into a vendor?
No. We design with abstraction layers so models, vector stores, and cloud providers are swappable.
What if the pilot doesn't work?
You get a full refund on the Starter Sprint if you're unhappy with the deliverable. For Growth Builds, we tie milestone payments to acceptance criteria.
Can you integrate with our existing tools?
Live today: Google Calendar, HubSpot, and Slack (self-serve OAuth), Shopify (read-only, manual setup), and Stripe (billing). Because the platform is API-based, we can connect most tools that expose an API — tell us your stack and we'll be honest about what's available now versus on our roadmap.
Who actually does the work?
Senior practitioners. Every client gets a named lead with 8+ years of experience. We don't pyramid junior hours.
Do you offer ongoing support?
Yes. Every Build comes with a 90-day care plan. After that, retainer plans start at $4K/mo.
What about compliance — GDPR and data handling?
We follow GDPR-minded data practices by default: data minimization, per-customer isolation, encryption, and deletion on request. Formal certifications (e.g. SOC 2, HIPAA) are on our roadmap, not yet held — if your engagement requires a specific certification or a signed agreement, contact us at info@thenexelgroup.com and we'll be straight with you about what we can support today.
Can we self-host?
On-prem and air-gapped deployments are available for custom enterprise engagements with strict data-residency or sovereignty needs. Contact us and we'll scope what's feasible for your requirements.
Still have questions?
A 20-minute call usually beats more reading.