Predictive maintenanceany machineany sensors

The data you already log knows what's about to break.

Sentys learns each machine's healthy behaviour from its own sensor history — no failure labels, no data-science project on your side — and tells you, in plain language, which machine is drifting and which sensor saw it first.

Send one export — get a free analysis See a sample report
TREND · ENGINE-03 · SURPRISE vs HEALTHY BASELINE MODEL OUTPUT — UNSEEN UNIT
warning · 3× healthy critical · 6× healthy healthy baseline · 122 quiet cycles first warning · cycle 123 — 56 cycles early failure · cycle 179

Real output, not an illustration: Engine 03 from the live demo below, exactly as Sentys scored it. Quiet for 122 cycles, first warning at cycle 123 — 56 cycles before it failed. Sentys had never seen this engine.

30/30
unseen engines flagged before failure
59
operating cycles of median warning
1
false-alarm engine across the fleet, thresholds set from healthy data alone
3
machine classes validated — one unmodified pipeline
LIVE — real model output on 30 unseen engines, not a mock-up

30 turbofan engines the model had never seen, run to failure. Every tile climbs green → amber → red as the machine degrades; click any engine for its per-sensor triage. The proof in plain terms: trained only on healthy operation of 70 other machines.

Open full-screen

What it is

Built to fit the plant you already have

No labeled failure history. No data-science team. No rip-and-replace. Sentys works with the sensor data your control system is already logging.

Learns healthy · flags drift

A baseline per machine, from its own history

It trains on your machine's own healthy operation — no failure labels, because at most sites they don't exist — and reports degradation as a rising deviation from that baseline, decomposed per sensor.

The answer is triage language, not an anomaly score: “Vibration: 3.2× healthy baseline; temperature and flow within range.”

Sensor-agnostic

Any tags your system logs

Temperatures, pressures, flow, current, vibration — it doesn't care. From a SCADA export, a historian, or a plain spreadsheet. New sensor types are configuration, not code — built by an integration engineer, so the data problem is solved first.

European exports parse natively: semicolons, decimal commas, your column names in your language. CSV or JSON, straight from the system you have.

Status · honest

Where it stands today

Today: an export is enough. Send one machine's history, get a written analysis back — zero integration on your side. A pilot: the same pipeline runs on-prem (Docker, no GPU) reading your exports or a live stream; your data never leaves the plant.

Not built yet: OPC-UA/Modbus connectors and hardened remote access. That's pilot-stage work we do together — we'd rather tell you now than surprise you later.

From signal to action

The drift becomes a work order — drafted for you

A rising trace is not an answer. Sentys's agents turn it into one: drafted by AI, grounded in your data, approved by your engineer. Every artifact below is real output from the pipeline, verbatim.

01 · PERCEIVE

The model listens

Every sensor stream is compared to that machine's own learned normal, every moment. Drift is reported per sense — rotation, temperature, flow — as a ratio to healthy.

machine · its own baseline signal · ratio vs healthy
02 · TRIAGE — DRAFTED BY AN AGENT

A work order, not a chart

WORK ORDER (DRAFT) — Inspect rotation subsystem on unit 100
severityescalate · confidence 0.6
dominant senserotation · 1.56× healthy
nearest prior caseunit 88 · kept degrading

“Schedule a rotation-domain inspection (bearing/shaft/rotor balance and vibration diagnostics) on unit 100 … compare findings against the near-identical prior unit 88 that continued degrading.”

Draft for your engineer to review — the tool proposes, your team approves. No automatic derate or shutdown.

Verbatim excerpt — benchmark unit 100, drafted by the triage agent, including its own stated caveats.

03 · THE WEEK AHEAD

A fleet report every week

Which machines need attention this week, ranked — the ranking is computed by code, never decided by the model. An agent writes the prose around it; a mechanical check rejects any draft citing a number that isn't in the evidence.

You read one page, not thirty dashboards.

Grounded or rejected: every number an agent writes is checked mechanically against the evidence it was given — a draft that cites anything the data doesn't support is rejected and redrafted. And nothing acts on a machine without a human signing off.

Evidence, translated

Three machine classes, one unmodified pipeline

Public benchmark datasets, trained on healthy data only, every time. Outcomes in plain language; the technical numbers sit in the small print for your advisor.

30/30 flagged before failure

Turbofan engines — NASA C-MAPSS

Every one of 30 engines the model had never seen was flagged before it failed — median 59 operating cycles of warning. With alarm thresholds set from healthy data alone, one engine in the fleet false-alarmed.

holdout: Spearman ρ = 0.762, AUC = 0.996 (FD001); ρ = 0.615, AUC = 0.95 on the six-regime FD002 subset — no regime-aware tuning.
75 h warning — and physics for free

Bearing vibration — NASA IMS

On run-to-failure rigs, the bearing that actually failed was ranked strongest-trending and loudest — flagged 75 hours before end of run on one rig, ~470 on the other. The spectral sense fired first, reproducing known bearing-defect physics with zero labels.

honest limits: same-shaft neighbours also elevate late (the ranked view still picks the right bearing by 2–3×); healthy baselines don't transfer across installations — each gets its own, which is the deployment protocol anyway.
±0.03 AUC of the specialist tool

Industrial pump audio — MIMII / DCASE

On four physically different real pumps in factory noise, the same pipeline — with zero audio-specific engineering — landed within 0.03 average AUC of the purpose-built DCASE audio baseline (0.687 vs 0.715), beating it outright on two of the four.

also measured: a brand-new pump gets useful day-one coverage from its fleet-mates' shared model (0.60 AUC unseen); its own baseline lifts that to 0.69 — cold-start from the fleet, precision from the fine-tune.

“How much warning do I actually get?”

The honest answer: on the turbofan benchmark, the median first warning came 59 cycles before failure — roughly the last fifth of the machine's life — and on the bearing rigs, days to weeks. What Sentys does not do yet is date the failure: it ranks which machine is drifting and shows which sensor saw it first, so you inspect the right asset while there's still time to schedule it. It doesn't output “fails in 12 days” — and we won't claim it does until we can prove it.

Everything above is public-benchmark evidence. We're seeking our first real-world pilot — that honesty is the point.

The deliverable

This is what you get back

Not a dashboard login. Not a sales deck. One self-contained report — readable by an operations lead, forwardable to the board.

  • Health-over-life chart for every machine in the data, with the warning point marked
  • Which sensor group saw the drift first, and how far from healthy it moved
  • Findings in plain language — and every caveat stated, in the same font size
  • Every number read mechanically from the run's artifacts — no hand-tuning, no cherry-picking
Open the sample report

Generated from a real analysis run of the NASA turbofan data — the same report generator your export goes through.

Where it fits

Any machine that hums, pumps, spins, or heats

Across Europe, the pattern is the same: the control system logs everything, and nobody reads the history. Sentys reads it. The machine classes it has been benchmarked on — pumps, bearings, turbomachinery — sit at the heart of every one of these.

District energy & utilities

The plant that keeps a town warm

Circulation pumps, heat exchangers, boilers — running unattended most hours of the year, where a pump failure in January is not a ticket, it's a cold town. Your control room already logs years of history at minute resolution. Sentys has never seen a district-heating plant — which is exactly why the first analysis is free.

Water & wastewater

Pump stations, around the clock

Submersible and dry-well pumps, blowers, screens — the benchmarked machine classes almost exactly. Drift in current draw or vibration shows up in the history long before the alarm bank knows anything is wrong.

Manufacturing & process

The compressor nobody thinks about

Compressors, fans, conveyors, CIP pumps — the machines that stop a line when they stop. One export per machine is enough to find out whether your last breakdown had warning in the data. Usually it did.

Robots & mobile fleets

Where Sentys was born

The pipeline started life as the nervous system of a real robot — learning the feel of its own motors and sensors, flagging what didn't feel right. Fleet predictive maintenance for robotics is where it's headed next; industrial plants are where it proves itself first.

Og ja — skriv gerne på dansk. Exports in any European locale parse natively — semicolons, decimal commas, and your column names in your own language. Reports come back in English or Danish, as you prefer.

Who's behind this

Frixos Andreou

Frixos Andreou

integration engineer · robotics background · Denmark, working across Europe

I studied robotics, and I've spent 5+ years as an integration specialist connecting enterprise systems — the unglamorous work of making data actually flow. Sentys started in my living room: a robot that learned the feel of its own motors and sensors from experience, with no labels, and noticed when something was off.

The same question — does this machine feel like itself today? — turned out to matter far more for a pump that keeps a town warm than for a robot. So that's what it does now. The benchmarks above are that one idea, tested honestly, three machine classes in a row.

LinkedIn frixos@sentys.ai

The ask — all of it

Send one export. Get a real answer.

Here is everything saying yes commits you to — all four steps of it.

  1. We sign an NDA first. You get a ready-to-sign mutual NDA before any data moves.
  2. You export one problem machine's history. The five-line spec is below — CSV or JSON, straight out of your control system. Twenty minutes, no project.
  3. Within a week you get the report. What drifted, which sensor saw it first, and whether your last breakdown had warning in the data. Free — one machine, no strings.
  4. That's it. If the report earns it, the next conversation is a 2-week shadow pilot on more machines — a separate decision, made on the evidence in your hands.

The data spec — five lines, in full

  1. One machine you care about — a pump, a heat exchanger, a motor.
  2. All its logged tags: temperatures, pressures, flow, current, vibration, speed.
  3. 1-minute averages are plenty; hourly still works.
  4. 6–12 months of history — ideally spanning a breakdown or repair you remember.
  5. CSV or JSON, any delimiter — semicolons and decimal commas are fine.

Open the full export spec (one printable page) — forward it to whoever runs your control system.

How your data is handled

  • Processed on a single machine in Denmark. Your raw data is never uploaded to any cloud service.
  • Stated up front: schema inference sends column names, per-column statistics, and a handful of sample values to an AI API — never your bulk data. If even that is too much, say so and it's done by hand.
  • Used for your analysis only. Never used to train models for anyone else. Deleted when you say so — confirmed in writing.
Email me — I'll reply within 48 hours
frixos@sentys.ai