AI process automation for operations teams
Vlad Tudor is an AI consultant and engineer in Bucharest who helps manufacturers, energy companies, distributors and back-office teams automate the manual processes that cost hours every week. He began as an industrial automation engineer, and his studio Sapio AI built an agent that screens public tenders against a company's profile and past projects.
Reply within 24 hours · no sales pitch
What I hear from operations teams.
“Everyone knows which process is manual. It's been on the list for two years.”
“We ran a pilot. It impressed the board and never ran on a Monday.”
“The data is there, somewhere between the ERP, a shared drive and Maria's spreadsheet.”
Rewire the process, then add the model.
When electric motors arrived, most factories simply replaced the steam engine that turned the central shaft, and for decades the gains were small. They came when factories were rebuilt around the task instead of around the power source. AI in operations is at the same point. Put a model on top of an unchanged process and you get a slightly faster version of the old one.
And AI amplifies whatever it is given: unclear rules and messy data produce confident, wrong output at speed. So the first weeks go on the process itself. Which steps disappear, which get checked by a person, and what data has to exist before anything is automated.
The full argument, from the REWIRED podcast.
What I'd do with you.
- 01
One manual process, measured before and after.
Invoices, orders, tenders, claims or delivery notes: the process that costs the most hours, taken from email and retyping to running with a person checking exceptions.
- 02
Monitoring that reads for you.
Tenders, supplier documents or regulatory updates, screened daily against criteria you write in plain language, delivered as a short digest.
- 03
Plant and maintenance knowledge that can be searched.
Manuals, maintenance logs, quality reports and shift notes, answerable by describing the problem, with the source document linked.
- 04
Working with the systems you have.
Most ERPs have an export, a database or a report to build on. Where the process knowledge can't go to a public cloud, the system can run on your own servers, isolated from the internet.
- 05
Training for the people who run it.
A workshop for the team whose work changes, so the exceptions are handled well.
AI workshops
What I've built.
Tender monitoring
A supplier bidding on public tenders was screening documentation on e-licitatie.ro by hand, where search covers only a few fields. Sapio built an agent that reads the detailed documentation of about 1,000 projects per run and scores each against a profile of the company, drawn from its past projects and previous work. In testing it picked out the tenders that fit, and it is now being prepared to run daily.
My own start was in operations
As an automation engineer I programmed PLCs, worked on backup-power systems and contributed to the control system for one of Romania's largest photovoltaic parks.
On-premise delivery
Sapio AI builds systems that run fully on the client's infrastructure, isolated from the internet, for data that must not leave.
Among European companies that considered AI and did not proceed, 70.3% say the reason is lack of relevant expertise. Cost comes much further down the list, at 38.4%.
In manufacturing, 44.7% of companies that considered AI name incompatibility with existing equipment and software as a reason they did not proceed, the highest of any sector.
Tell me which process your team would most like to stop doing by hand. We'll measure it first.
A few short questions. The first conversation is free.
Reply within 24 hours · no sales pitch
What an engagement can cover.
A first conversation
About the problem, before any technology. Free.
An audit of processes, data and tools
How the work runs today, what data exists, which tools are already paid for.
A prioritised roadmap
What to build, what to buy and what to skip, in order, with the reasoning written down.
The build, or a clean hand-off
Hands-on build through Sapio AI, or a roadmap your own team can take from there.
Team enablement
A workshop for the people who will use the result, so it gets used.
Follow-up
A check on what changed once the work is in place, and what comes next.
For operations: the audit happens on site, where the work happens (plant, warehouse, accounting), on real extracts from the ERP rather than a questionnaire; the roadmap names the systems and who grants access.
Questions that come up.
Does it work with our ERP?
That's the first thing the audit checks, on your real data rather than a questionnaire. Most ERPs have an import path, a database or a report that can be used. Where there is no API, the work starts from what exists.
What happens with the exceptions?
They're part of the design. The system routes what it can't handle to a person, with the reason, and the share of documents that go through untouched is how we measure it.
Can the data stay on our premises?
Yes. Systems can run entirely on your own servers, isolated from the internet, when the data can't leave.
We already tried an automation tool and it broke. Why would this be different?
Often it broke on the first exception because nobody wrote down the rules the team follows in their heads. The audit writes them down first. If the process still can't be automated reliably, you'll hear that before anything is built.
Do you work outside Bucharest?
Yes. The audit is best done where the work happens, so I come on site for it; the rest runs remotely.
Tell me what you're working on.
A few short questions, two or three minutes. The reply comes from me, within 24 hours on working days.
Reply within 24 hours · no sales pitch
Help with
What would you like help with?
Pick all that apply.