AI should make work better for the people doing it.
That is the bet behind AgentLabs, a compact agentic transformation studio: we accelerate workforces with agentic systems where agents carry the volume, people keep the judgment, and your team owns the result. Led by Bram van Gestel, twenty-plus years scaling enterprise SaaS, high-demand consumer platforms, and recruitment operations as a hands-on CTO, CDO, and Director of Engineering, plus two ventures built from the ground up. No layers between strategy and implementation.
What we believe.
Five convictions that shape every engagement, before a single tool is chosen.
- 01
We redesign the work before we automate it.
Most AI projects bolt tools onto a process that was already broken. We map how the work actually runs, its handoffs, exceptions, and cost, then design the agentic workflow and the human checkpoints. Only then do we build.
- 02
Documentation and handover are deliverables.
A system that only runs while its builder is in the room is a demo, not a transformation. We write the operating model down and hand it over, so your team can run, understand, and improve the system after we leave.
- 03
Cost and value stay visible from day one.
You see what a workflow costs to run and what it returns, in the open, from the first week. Spend accountability is part of governance: no black-box spend, and no transformation nobody can point a number at.
- 04
A person stays in the loop where judgment matters.
We build human-orchestrated systems: agents carry the volume, a person owns the decisions that carry risk, by construction, not by convention. Where that line sits is a deliberate design choice, and it is where most of the work is.
- 05
Gates, not guidelines.
A rule the system enforces protects you; a rule in a policy document hopes for the best. We build guardrails as gates: what an agent can never touch, what cannot ship without sign-off, what is always logged. The gates make the work safe to review, and a person still decides what ships.
Every system we hand over has the same shape.
Agents carry the volume. A person owns the decisions that carry risk. Your team owns the result. We call that shape the Handover Architecture.
Agents do the work
Research, drafting, routing, reconciliation: the high-volume execution.
- human decides
A person decides
The calls that carry risk or judgment stay with a human.
- you own it
Your team owns it
Documented and handed over. Yours to run, and to improve.
Systems that survive handover.
A transformation is measured by what your team can run once we leave. Every system is shaped to be owned, not just delivered.
- Maintainable by design
- Documentation as a deliverable
- An audit trail of what agents did
- Cost visibility from day one
- Measurable value
- An operating model your team runs
Work directly with the people who build it.
No handoff between the strategist who sold it and the team who builds it. Bring a workflow and find out what that looks like in practice.
No sales pressure. You get an honest read on your workflow, including where AI does not help.