Fewer manual tasks.
Processes that work.
I automate repetitive work and connect the tools you already use, with clear rules, traceability and human review. You do not need to know which metric to improve: we measure it and agree it before we start.
Python · SQL · APIs · ETL · language models · voice
How the solutions work
Every flow separates the deterministic from the probabilistic: clear rules for what can be verified, AI only where it adds value, and a person deciding on the cases that matter.
- 01
Input
Documents, spreadsheets, messages or calls that are handled by hand today.
- 02
Deterministic rules
Validation, transformation and ETL in code: reproducible, auditable results.
- 03
Scoped AI
Model-assisted extraction and classification, with explicit limits and error handling.
- 04
Human review
Doubtful or sensitive cases are flagged and decided by a person.
- 05
Traceable output
Records, exports and notifications with a history of what happened.
Processes that can be automated
You do not need to fit a sector: tell me which process takes up your time. These are four common examples.
Accounting / bookkeeping
Invoices and documents
The team stops re-entering data and errors are caught before they reach the system.
Documents and data
Consultancy / agency
Integrations and technical capacity
Extra capacity for a peak of work, with documented delivery.
Integrations between tools
Gym / clinic
Calls and appointments
Calls outside opening hours or at peak times are recorded and handed over instead of being lost.
AI voice agent
Academy / school
Lead capture chat
Repeated questions are answered and requests are recorded, on the channel your customers already use.
AI in existing processes
How to start
No open-ended projects or blind quotes.
- 01
Conversation
A free 30-minute call to see whether the case fits. Before that, a message with the context.
- 02
Diagnosis
Indicative €100–180: process map, baseline and a go/no-go recommendation.
- 03
Pilot
What changes, timeline, condition and a fixed price. Ongoing improvement only if the flow needs it.
The video demo
A chat assistant that answers, qualifies and books a call. The video shows the full flow with synthetic data.
No sign-up and no data sent. If you want to see how it would fit your process, let’s talk.
Video in Spanish
Watch the demoYou do not need to know which metric to improve
In the diagnosis we measure how the process works today and agree what is worth moving: time, errors, cost, capacity or missed calls.
Before
Volume, time per case, errors and current cost.
Target
What changes and how it will be checked, in writing.
After
A report with what was processed, what needed review and what was improved.
Two questions before we start
Better to say them now than later.
- Can I not just do this myself with Claude or a no-code tool?
- Sometimes you can. If the process is simple and does not touch critical systems, I will tell you before selling you a project. Doing it yourself is usually not the right option when data lives in another system, nothing can be lost, an audit trail is needed or something has to respond when it fails in production.
- Are you going to replace my team?
- No. I automate the repetitive part so your team can focus on what needs judgement. In sensitive cases, the system flags and hands over; it does not decide.
Do you have a process that takes up time?
Tell me how it works today: frequency, volume, tools and who decides. If you do not know which metric to improve, we will work it out together. If there is a clear case, I will propose a diagnosis and a scoped pilot with a price and defined criteria.