LISBON · 2026

Applied AI agents on real data, built and validated in weeks.

We're a studio of engineers who work inside the client's operations to build functioning AI agents on real data, from diagnosis to prototype in weeks.

Schedule a meeting

Trusted by

Teleperformance
Semapa
Immersis
CTT

We build AI agents. Fast. Secure.

  1. Fast to start.

    We build inside the Microsoft 365 you already use. No new platform means no security review and no IT permissions to wait for.

  2. Fast to build.

    A first working agent on your own data in one week. You see it work before you commit to anything.

  3. Fast to go live.

    The prototype already runs in your real systems, so going to production means hardening it, not rebuilding it.

  4. Secure from day one.

    Your data never leaves your tenant, and access follows the permissions you already have.

  5. Built to be used.

    The people who will use the agent help us build it. By launch day they already know it and trust it.

Forward Deployed Engineers.

DIAGNOSIS AND FIRST PROTOTYPE1 week

In one week we diagnose the problem and deliver a first working prototype, running inside your systems on your real data.

CHANGE MANAGEMENTDay 1

Our engineers work inside your team, next to the people who will use the agent. They shape it with us from the first day, so adoption starts long before launch.

The engineers who design the agent are the same ones who build it.

Three stages in sequence, each one validating the decision to continue.

1

On-Site Diagnosis

≈ 1 week on-site

Engineers on-site, direct access to real systems. Deliverable: mapping, architecture and phased plan.

2

Prototyping Cycle

≈ 1 week

Functioning prototype on real data with validated return before any production commitment.

3

Production Implementation

Ongoing program

Agents in production with measurable return and ownership transferred to the team.

Return from day one, risk spread across each stage, no commitment before evidence.

Different problems. Same way of working.

~/agents/pharmacy-analyst

Healthcare software company

Before
Consultants spent their days collecting and formatting data from scattered systems. It all depended on undocumented Excel macros and the one person who maintained them.
The agent
Collects and cleans the data on its own, drafts the analysis and the report for each pharmacy, and keeps a record of every decision. The consultant reviews and decides.Built entirely inside the client's Microsoft 365. No pharmacy data leaves their environment.
Result
The freed-up capacity turned into more than €1M in additional projects delivered.

A senior team, embedded in your operation.

Miguel Mesquita da Cunha

Miguel Mesquita da Cunha

Founder

Aerospace engineering and computer science at IST. Previously in private equity and at a European mobility scale-up.

Augusto Vasco Costa

Augusto Vasco Costa

COO

Strategy and operations experience across startups and venture capital.

Let's build.

Let's talk about your first agent. No commitment, no generic deck.

Open the calendar

miguel.cunha@waldyn.eu · waldyn.eu · +351 918 056 995