Frequently asked questions.

We design and build AI systems that perform business work: AI employees for specific roles, agentic workflows, automation, knowledge systems, and the web applications and SaaS products around them — from strategy through production and ongoing improvement.

A supervised AI system built around one business function, such as sales development or support triage. It has defined responsibilities, access to specific tools, a knowledge base, and rules for when to hand work to a person.

Not by default. Agents launch in supervised mode, and specific low-risk actions become autonomous only after they have proven reliable. Sensitive and high-impact decisions stay with people.

No. We are a complete delivery partner: strategy, product design, web and SaaS engineering, integrations, and AI. Many projects are mostly software with AI where it adds clear value.

Least-privilege access, scoped credentials, approval gates, audit logs, environment separation, data minimisation, evaluation before release, monitoring, and the ability to pause or roll back any agent.

With a strategy call about one workflow or product idea. If there is a fit, we propose a discovery phase or a focused first build with a clear scope.

A supervised AI system built around one business role. It has defined responsibilities, access to specific tools, a knowledge base, and rules for when to involve a person.

No. It works autonomously only on tasks that have proven reliable in supervised operation. Sensitive and high-impact decisions remain with people.

Roles with a high volume of repetitive, rule-based work spread across digital systems — sales development, support triage, operations coordination, and back-office finance are common starting points.

Through a review queue, an activity log, and regular performance reviews with our team. Managing it should feel closer to supervising a colleague than maintaining software.

An agent is the technical pattern — a system that uses tools in a loop to complete tasks. An AI employee is one or more agents packaged around a business role, with knowledge, supervision, and reporting.

They can be, when access is scoped, actions are logged, consequential steps require approval, and behaviour is tested before release. Safety comes from engineering, not from the model alone.

Yes, multi-agent workflows are possible. We only use them when they make a system simpler to reason about — not by default.

Not at first. Agents launch in supervised mode where actions are reviewed. Specific low-risk actions move to unattended execution only after they have performed reliably against agreed criteria.

Errors are expected and designed for. Uncertain cases are escalated, actions are logged, high-impact steps require approval, and evaluation suites catch regressions before release.

Strategy work usually takes a few weeks. A focused first build is typically measured in weeks to a few months, depending on integrations. We give a timeline after discovery rather than before.

Discovery and strategy are usually fixed-scope. Builds are scoped and priced per phase. Ongoing engineering is a monthly engagement. Details are agreed in a proposal.

Ownership is set in contract. Our standard approach is that clients own the code and configuration we build for them, in repositories and environments they control.

Yes. We work with clients internationally and agree meeting times and communication channels at the start of each engagement.