The AI Experimentby Xelec

Company

Small, senior, and allergic to the pilot graveyard.

The AI Experiment is Xelec's AI venture. We take a small number of engagements at a time and staff them with people who have already put this class of system into production.

Principles

Four commitments we make in writing.

A production gate before the work starts

We agree in writing what 'this works' means — accuracy, latency, straight-through rate, cost per unit — before the first sprint. Pass it and we ship. Fail it and you keep the findings.

Senior people, no pyramid

The engineers on your call are the engineers writing the code. We do not staff a partner over three juniors and bill the difference.

Your data trains nothing of ours

No shared training, no derived-model reuse, no clause hiding in the appendix. You own the work product and any model trained on your data.

We will tell you not to build it

Roughly one in four discovery calls ends with us recommending against the project or a cheaper non-AI fix. That is the point of talking to engineers.

Delivery model

Built for US and European buyers specifically.

Time zones, procurement expectations, residency rules and regulatory obligations differ, and generic offshore delivery models ignore all four. Ours does not.

  • Overlapping hours

    Daily overlap with both US Central and Central European working hours.

  • Procurement-ready

    DPA, security questionnaire responses, sub-processor list and insurance documentation on request.

  • Regional delivery

    Deployment and data handling in your region, with named engineers accountable for it.

  • Regulatory literacy

    EU AI Act classification and GDPR obligations treated as design inputs, not paperwork after the fact.

Engagements

Four ways to work together.

Discovery sprint

2 weeks

Fixed fee

Use-case scoring, data reality check, costed sequence. Credited against a pilot if you proceed.

Paid pilot

4–8 weeks

Fixed scope, fixed fee

One use case taken to a written production gate, with the evaluation harness you keep either way.

Build partnership

Quarterly

Embedded team

A senior team embedded against a roadmap: production systems, hand-off runbooks, internal capability transfer.

Managed operation

Annual

SLA-based

We run and monitor what we built — drift, cost, guardrails, upgrades — against an agreed SLA.

Assurance

The answers procurement asks for.

EU and US regions
Data residency
Your cloud, our cloud, or on-prem
Deployment
EU AI Act and GDPR aligned
Regulation
SOC 2 aligned practices
Controls
Model-agnostic, no lock-in
Model policy
You own the work product
IP

Next step

Start with a conversation, not a contract.

Thirty minutes with an engineer who has shipped this before. If there is no fit, we will say so on the call.