Digitalisation for industry · Edge AI · Computer vision · Athens and Nicosia

We digitalise the work your factory still does by hand.

Plaixus turns the documents you already issue and the lines you already run into records you can act on. Software, together with the tags, tablets, cameras and edge devices it needs on the floor. No change to your ERP, nothing wired into the machine.

Two products are in the field today. When your job is different, the same team builds it: a two-week data audit first, with a written go/no-go before you commit to anything.

Two productsTraceLM for material certificates, LatentML for line performance
Software with hardwarecameras, encoders, tags, tablets and edge devices, installed with Robotico
Passport-readyrecords built for the EU Digital Product Passport, with steel among the first product groups
In productionTraceLM issues EN 10204 certificates daily at a listed steel producer

Products

Digitalisation, ready-made

Each product makes one part of the factory's record digital. Each one started as a job a customer asked us to do, and now runs as a service for a monthly fee.

TraceLM

Material certificates, issued from your own files

In production · EN 10204 3.1 & 2.2

TraceLM reads your mill certificates, links each delivery to its coil and heat, and issues the EN 10204 certificate. It works on the files you already have, and it runs in production today at a Greek steel producer.

TraceLM is live. It issues 2.2 test reports and 3.1 inspection certificates to EN 10204 for real deliveries, from the producer’s own invoices, mill certificates and tensile tests.

A Digital Product Passport plugin for the systems you already have: DPP-ready, no ERP change.

TraceLM screen with an inspection certificate 3.1 to EN 10204, filled in from mill data and a tensile test
An inspection certificate 3.1, filled in by TraceLM. From the installation in production.
  1. CollectFiles sync from your own PCs: invoices, delivery notes, mill certificates and tensile tests.
  2. ReadAI reads each mill certificate, digital, scanned or faxed, and finds every heat with its chemistry and mechanical values.
  3. LinkTraceLM matches each invoice line to your receipts and coil records: the coil, its heat and its piece number.
  4. IssueThe right certificate for every delivery: a 2.2, or a 3.1 with your own tensile test.
TraceLM in detail Start with a pilot

A recurring monthly fee. No change to your ERP.

LatentML

Line performance from a camera or an encoder

Pilots open · shown on a working line

LatentML measures Overall Equipment Effectiveness (OEE) on any machine, from an IP camera pointed at the line or a clip-on encoder on a shaft or roller.

It counts cycles, finds stops and slow running, and shows real-time OEE per line and shift. There is no PLC integration, no relay or sensor in the cabinet, and nothing for the machine builder to do. The models run on an industrial mini-PC at the line, and the video never leaves the factory.

The film shows it on a working line in electrical-waste recycling. The customer remains confidential.

LatentML film frame: a camera view of a conveyor with every device boxed and numbered, and the line status shown as running
Every device counted, and the line status read from the camera. From the film.
  1. InstallAn IP camera at the line or an encoder on a shaft. Half a day, with no stoppage.
  2. TeachYou mark the counting zone on screen. Thirty minutes of normal production.
  3. RunThe edge device counts and classifies line states on site, around the clock.
  4. ReportDashboards, a Pareto of losses, shift reports and stop reasons from the operator.
LatentML in detail Start with a pilot

A subscription per line, per month, with the edge device included.

How we think

The factory already produces the data. It sits on paper and in people's heads.

We do not start from a new system. We start from what a factory already does, and make it readable.

Start from what exists

The documents you already issue and the lines you already run. Your ERP stays as it is, and nobody types anything twice.

Software with hardware

A record starts on the floor, so the software arrives with what it needs there. Robotico installs and supports it.

  • Cameras
  • Encoders
  • Tags and label printers
  • Tablets and scanners
  • Edge devices

Records first, automation next

Once every certificate, line and order has a record, machines can act on it. A Digital Product Passport is built from the same records.

A worked example

The paperwork already knows. Nobody has time to read it.

Goods leave for a subcontractor, and the order drops out of sight until they come back. Twenty slides follow one such problem from the dispatch note to the invoice, in eleven steps, with the scans, tags, tablets and cameras each step needs.

Open the deck
OrderNote outNote backInvoice

Why now · Digital Product Passport

The next wave is a passport for every product

The EU is bringing in a digital record that travels with each product. Nobody can produce one from paper, spreadsheets and memory. Digitalising now is how a factory arrives ready.

  1. April 2025The EU names iron and steel a priority product group under its ecodesign rules (ESPR).
  2. 2026Steel-specific rules expected, setting what the passport must contain.
  3. Around 2028Digital Product Passports expected for steel products sold in the EU.

A passport draws on the product's record. We build that record one part at a time.

  • What it is made ofGrade, composition, mechanical properties and the EN 10204 certificate, traced from heat to delivery. TraceLM holds these today.
  • How the line ranWhat each line produced, and when it ran, stopped or waited. LatentML measures this at the line.
  • Where each order standsWhich order every item belongs to, and who is holding it. The deck above follows one such case.

Start with steel

Steel is among the first product groups. TraceLM carries a passport plugin for the systems you already have, with no ERP change.

Steel comes first. In two funded projects we already write passport records for other products.

REVIVE · e-mobility recycling

A passport entry for every device model taken apart

At a recycling plant, a robot cell takes e-scooters and hoverboards apart for reusable parts. For each device model we write the passport entry: materials, hazards, the value of each component and where it should go next.

REVIVE, with Bendabl and HRC

PROTEUS · EV battery disassembly

The disassembly sequence, recorded once and kept

On the bench, each step an operator takes on an unfamiliar battery is recorded and stored in the product's passport. The next operator, or a robot, follows the same sequence.

PROTEUS, led by Robotico

Source: European Commission ESPR working plan, 16 April 2025. Dates are indicative.

Who we are

A deep-tech team that ships models, not slide decks

Plaixus is a Greek deep-tech SME founded in 2022 by an engineering team with research roots. We specialise in industrial-grade machine learning: computer vision and edge AI, MLOps, explainable models, and language models applied to industrial documents. On our platforms or yours.

Most AI projects stall between a promising prototype and a system operators can trust. That gap is our whole business. We work across the full lifecycle: data ingestion and curation, model training and optimisation, edge deployment, monitoring and explainability.

The name comes from plexus: a network of pathways that carries signals between the centre and the extremities. That is what we build: code and data moving gracefully along the cloud-edge continuum.

Book a discovery call
  • Production firstEvery model is built to be deployed, versioned, monitored and retrained, not just evaluated.
  • Runs on what you haveCameras, PLCs, SCADA, ROS robots, nanosatellites: we integrate rather than replace.
  • Privacy and compliance by designOn-premise pipelines and federated learning when data cannot travel.
  • Fast, honest feasibilityA two-week data audit tells you whether the problem is solvable before you commit.

What we build

The same work, made to order

When your job is different from the products, the same team builds it. These six capabilities are what the products are made of, and most engagements end with a monitored pipeline running at the edge.

Computer vision & edge AI

Detection, tracking, counting, anomaly and quality inspection models that run on-device with low latency and no dependency on a stable uplink. Our livestock lameness-detection work under the AGRARIAN programme is a typical example: cameras in the barn, inference at the edge, alerts to the farmer.

  • Object detection & segmentation
  • Multi-camera tracking
  • Visual anomaly detection
  • Quantisation, distillation & pruning
  • NVIDIA Jetson & industrial gateways
  • Offline-first inference

LLM document & telemetry intelligence

Language models applied to documents and machine data, not chatbots: structured extraction from certificates and test reports, retrieval over your own technical documentation, and translation of analyst questions into detection queries over security telemetry. Every output carries a confidence score and goes through human review. Runs on-premise on open-weight models when the data cannot leave. TraceLM is built this way.

  • OCR + LLM extraction
  • Retrieval-augmented generation
  • Tool-using LLM workflows
  • On-premise / open-weight LLMs
  • Evaluation & guardrails
  • Confidence scoring & human review

MLOps & AI pipelines

Reproducible training, automated deployment and continuous monitoring so a model stays accurate long after launch. Drift alerts, retraining triggers and explainability reports your operators can read.

  • Kubeflow
  • Argo Workflows
  • MLflow
  • Model registries
  • Drift & performance monitoring

Federated & privacy-preserving learning

Train shared models across hospitals, homes or plants without moving raw data. Full compliance with data-protection rules, with the accuracy of a pooled dataset.

  • TensorFlow Federated
  • PySyft
  • Differential privacy
  • Secure aggregation

Cloud-native & IIoT integration

We connect models to the systems that matter: DCS, PLC and SCADA on the plant floor, ROS 2 on robots, ESA's NanoSat MO Framework in orbit, with Kubernetes orchestration end to end.

  • KubeEdge
  • EdgeX Foundry
  • Fledge
  • ROS 2
  • OPC UA / Modbus

Data analysis & decision support

From sensor time series to dashboards and forecasts: predictive maintenance, yield estimation, grid optimisation and explainable models (SHAP, LRP, LIME) your operators can audit.

  • Time-series forecasting
  • Predictive maintenance
  • NLP & document intelligence
  • Explainable AI (XAI)
Book a discovery call Not sure which capability applies? Describe the data you have and we will tell you.

How it fits together

One pipeline across the cloud-edge continuum

Training, optimisation and orchestration happen in the cloud or your data centre. Inference, buffering and control happen at the edge, next to the camera, the controller or the payload computer. Plaixus is the layer in between that keeps the two in sync: models flow down, telemetry and drift signals flow back up.

Because the layer sits on open frameworks, you are never locked into our tooling, and your team can operate it after handover.

Book a discovery call
Cloud-edge continuum diagram Cloud training and orchestration at the top, edge devices at the bottom, with the Plaixus MLOps layer in the middle. Models flow down to the edge; telemetry flows up. Cloud / data centre Training Optimisation Registry Orchestration Plaixus MLOps layer Deploy · Version · Monitor · Explain · Retrain KubeEdge · Kubeflow EdgeX · Argo · ROS 2 models telemetry Edge Cameras PLC / SCADA Robots Satellites
Navy paths carry models down; cyan paths carry telemetry, drift signals and labels back up.

Platforms

Reference platforms we deploy and adapt

Each one packages a proven pipeline for a specific environment. We use them to start your project weeks ahead, then tailor the last mile to your equipment and data.

  • AIstronaut

    Space · edge AI in orbit

    Deploy and monitor edge AI applications on satellites, combining the power of the cloud with the timeliness of on-board inference for Earth observation and payload autonomy.

    ESA NanoSat MO Framework, KubeEdge, EdgeX, Kubeflow
  • RoboSense

    Robotics & edge fleets

    Kubeflow and K3s adapted for the realities of the edge, so the same platform runs in a data centre or on the device. Two ways in: build and train models on the platform, or bring a model you trained elsewhere and ship it to any edge device you connect to it. Inference on the device is optimised through NVIDIA Triton.

    Kubeflow, K3s, NVIDIA Triton, ROS 2, Argo
  • Pledge

    Manufacturing & process industry

    Bridge industrial IoT with the cloud and your existing DCS, PLC and SCADA. Removes the data silos that block AI applications in plants and factories.

    KubeEdge, EdgeX, Fledge, OPC UA
  • Helianthus

    Energy & smart buildings

    Train federated models on smart-home and grid data streams. Extends to distribution system operators for substation and RTU data, optimised grid use and predictive maintenance.

    Home Assistant, openHAB, HomeEdge
  • FederatAId

    Healthcare

    Use multiple healthcare data repositories to train federated AI models in full compliance with data-privacy regulations, without centralising patient data.

    Kubeflow, KubeEdge, TensorFlow Federated, PySyft

Sectors

Where our models are running

Physical environments and regulated processes with real constraints: dust, distance, latency, compliance. That is where applied AI pays for itself.

Manufacturing & process industry

Visual quality inspection, predictive maintenance on rotating equipment, and IIoT data unification across legacy control systems.

Steel & metals

Mill test certificates read automatically, EN 10204 3.1 and 2.2 certificates issued with full traceability, and Digital Product Passports ready for ESPR (TraceLM).

Agriculture & livestock

Camera-based animal health monitoring, yield and growth estimation, and offline-capable field devices for farms with weak connectivity (livestock monitoring).

Robotics & autonomous systems

Perception models integrated with ROS 2, deployed and updated over the air to fleets of mobile or industrial robots.

Energy & utilities

Federated analytics on grid and building data, predictive maintenance for substations, and forecasting for optimised grid use.

Space & Earth observation

On-board inference for nanosatellites, so useful insight comes down instead of raw imagery that saturates the downlink.

Cyber security

Language models that translate analyst questions into detection queries over SIEM and IDS telemetry, mapped to MITRE ATT&CK, with the generated query shown for review.

Healthcare

Federated model training across institutions, keeping patient data where it is while improving shared diagnostic models.

Book a discovery call Your sector is not listed? If the data comes from cameras, PLCs, sensors or certificates, it probably fits. Ask.

Research & funded projects

Research-grade ideas, field-grade delivery

We participate in European R&D as a technical partner that turns concepts into deployable systems. If you are assembling a consortium and need edge AI, MLOps or integration expertise, we would like to hear from you.

Propose a collaboration

The Horizon Europe projects listed here are carried by Plaixus Limited, our sister company in Cyprus, which takes part in EU-funded research. All projects and publications.

Research interests

  • Federated & gossip learning
  • Delta-sync edge computing
  • Quantisation, distillation, pruning
  • Conditional computation: MoEs, token selection, early exit
  • Vision-language models (VLMs)
  • Audio-visual fusion
  • Spiking neural networks
  • Agentic frameworks: reflection, tool use, planning, multi-agent
  • XAI: LRP, SHAP, LIME
  • Causal & neuro-symbolic AI
  • Neuromorphic computing & SNNs
ARISE · Horizon Europe · 2024 to 2027

Robots that learn farm and solar-site tasks

Robotic arms with grippers, perception and learning tools for jobs such as installation, repair, transplanting and harvesting. Coordinated by Tech Hive Labs. Our part is the edge AI platform the partners register and deploy their models through, on workstations and on edge devices, and the compression that makes object detection fit those devices. Published so far: CargoSync, which updates software on edge devices over low-bandwidth links. Project record on CORDIS.

MIRANDA · Horizon Europe · 2024 to 2027

A digital twin for responding to cyber-attacks

A digital twin of interconnected services that helps organisations detect and respond to cyber-attacks across their supply chains and the computing continuum. Our task is the integration pipeline, which builds each partner's component and deploys it to a shared cluster. Project record on CORDIS.

SHARC · Horizon Europe · 2026 to 2029

6G radio networks on shared, accelerated hardware

Energy-efficient, flexible and secure 6G radio access, with workloads moved across mixed hardware accelerators. Plaixus Limited leads the work package on impact: communication, standardisation and exploitation. Project record on CORDIS.

REVIVE · ROB4GREEN open call · 2026 to 2027

A robot cell that harvests parts from end-of-life e-scooters

REVIVE is building a collaborative robot workstation at the e-waste plant of HRC (Elliniko Kentro Anakyklosis) in Greece. It takes light e-mobility devices such as e-scooters and hoverboards apart for reusable parts instead of shredding them. Bendabl coordinates and builds the robot cell. We lead the AI and software: the vision that identifies each component, a digital twin of each device model, and a Digital Product Passport entry per model.

PROTEUS · AID4SME open call · with Robotico

Work instructions projected onto the bench for EV battery disassembly

Selected in the second AID4SME open call. Projectors and depth cameras above the bench will record how an operator dismantles an EV battery or motor the plant has not seen before, then replay those steps as instructions projected onto the bench and the product. Each operation, fastener, tool and hazard goes into a Digital Product Passport, so the sequence can guide the next operator or a robot. Robotico leads, with Plaixus Limited as partner.

AGRARIAN · cascade funding

ELDER: Edge-AI for Lameness Detection in Ruminants

Computer vision that scores individual sheep for lameness from video in the barn. Each animal is tracked across a sequence of frames and the sequence produces one probability, rather than a verdict per frame, with inference running on-device. What it measured in the field.

SHEPHERD-AI · O-CEI open call · 2026

Fever and milk-drop alerts for dairy cows

A thermal camera reads the temperature at each cow's eye and flags a likely fever. A second check flags any cow whose daily milk yield falls more than 15% below her seven-day average. Detection runs on edge nodes at the farm and is designed to raise an alert within five minutes. The health data follows a shared schema, so other applications on the O-CEI platform can use it. Started in March 2026.

SPIRE · I-NERGY cascade funding

Edge AI for photovoltaic predictive maintenance

An open-source edge AI platform built as an add-on to SAV3, ELVAN's cleaning robot for solar panels. Image and vibration models on the robot detect soiling and loosened bolts, running on-device at the array. Selected as one of 15 funded projects from 80 applications across 26 member states in the I-NERGY second open call. What it taught us about intermittent links.

NAVIR · dAIEDGE

Neuromorphic audio-visual interaction for robotics

Audio-visual speech recognition running as spiking neural networks on a neuromorphic accelerator, driving a robot arm from a single-board computer. One of the first demonstrations of AVSR on this class of processor. Measured energy and accuracy.

Publications

What the team has published

Papers with Plaixus authors. Several are joint work with Tech Hive Labs and the AI Innovation Centre of the University of Essex.

Earlier work by the team at TWI Hellas, now Tech Hive Labs, includes a decision support system for prosthetic socket fitting (Sensors, 2021). Tech Hive Labs keeps its own list at thlabs.eu/publications.

Who we work with

Digitalising a factory takes more than software

These are the companies we build, install and research with. One team arrives on site.

Machines, installation and support

Robotico

Designs and builds dedicated machines for repetitive factory work, and installs and supports TraceLM and LatentML on the floor.

robotico.gr

Networks and connectivity

iT-TechLab

Structured cabling, Wi-Fi, 4G and 5G failover, satellite internet and IP cameras: the network a digital factory runs on, designed, installed and supported across Greece.

ittechlab.gr

Applied research

Tech Hive Labs

Non-profit organisation for applied research in robotics and AI. Coordinates the Horizon Europe project ARISE.

thlabs.eu

EU-funded research

Plaixus Limited

Our sister company in Cyprus takes part in Horizon Europe projects, among them ARISE, MIRANDA and SHARC, and in cascade-funded sub-grants.

R&D

How we work

From first call to a monitored system in four steps

Fixed scope at each stage, so you decide with evidence rather than optimism.

Discovery call

We hear the operational problem, look at sample data and agree what success would look like in numbers.

45 minutes, free

Data & feasibility audit

We assess data quality, edge constraints and integration points, and return a written go / no-go with an architecture sketch.

About 2 weeks

Pilot on your site

A working model on one line, herd, robot or site, deployed on real hardware and measured against the agreed targets.

8-12 weeks

Scale & operate

Roll-out across sites with monitoring, retraining and handover documentation, so your team owns the system.

Ongoing, as needed

FAQ

Questions we hear on the first call

How long does an edge AI pilot take?

A scoped pilot on your own data typically runs 8 to 12 weeks: two weeks for a data and feasibility audit, then iterative model development and a field deployment on one line, herd, robot or site.

Do we need new hardware?

Usually not. We act as an intelligence layer on top of the cameras, PLCs, sensors and controllers you already run. When an edge device is needed we recommend commodity hardware such as NVIDIA Jetson or industrial x86 gateways.

Can our data stay on-premise?

Yes. Our pipelines run on your infrastructure or a private cloud, and for multi-site or sensitive data we use federated learning so raw data never leaves its source.

Do you join EU-funded consortia?

Yes. We have delivered cascade-funded and Horizon Europe work as a technical partner, contributing edge AI, MLOps and integration expertise, and we co-write proposals with partners.

What is TraceLM and who is it for?

TraceLM is our platform for steel processors and traders. It reads mill test certificates and tensile-test PDFs with OCR and LLM extraction, validates the data, and issues EN 10204 3.1 and 2.2 certificates with full traceability from mill to customer. A Digital Product Passport plugin prepares products for the EU ESPR rules. It runs in production at an Athens Stock Exchange-listed steel company.

Where do you work?

We are based in Athens, Greece, with a second office in Nicosia, Cyprus, through our sister company Plaixus Limited. We work with clients and consortia across the European Union, on-site when needed and remotely by default.

Book a discovery call Your question is not here? It will be answered on the call, in writing afterwards.

Contact

Tell us what your data should be doing

Describe the problem in a few lines. An engineer, not a sales rep, replies within one business day with next steps or honest reasons why AI is not the right tool.

Based in
Athens, Greece and Nicosia, Cyprus · working across the EU
Hours
Monday-Friday, 09:00-18:00 EET
Elsewhere
LinkedIn · GitHub

We reply within one business day. Your details are used only to answer your enquiry. Privacy.