Your plant runs.Your time is yours.

We are building AI agents that keep watch over an industrial plant, inside limits its owner signs. A fault at two in the morning becomes one tap on your phone, not a drive to the plant.

Pre-seed, from Managua, Nicaragua. A working prototype runs on a digital twin of a reverse-osmosis water plant. We are preparing the first pilot on a real plant.

Lux Cognition02:14 · RO plant

The membranes need cleaning

Normalized flow is down 13.9% since the last clean and the permeate tank is at 58%.

Proposed by the process agent

Start a chemical clean (8 hours). It needs your approval: it pauses water delivery to the mine.

  • Inside the limits you signed
  • The instruments it depends on are healthy
  • Written to the tamper-evident log

Try it. Synthetic data from our digital twin.

The plant runs all night. You shouldn't.

People who own a plant tell us the same thing. They want more time for family, for travel, for the next business. What stops them is the fear of what happens when they are not there.

  • The call in the middle of the night

    Something drifts, an alarm goes off, and the decision waits until the owner picks up the phone.

  • Every shift runs the plant differently

    The data is already there, in the PLC and the SCADA. What happens with it depends on who is on shift.

  • Being away feels like a risk

    So the owner stays close. Lux Cognition exists to change that: the plant is watched, and the owner decides only what truly needs them.

Agents on watch, inside your limits

Each agent looks after one part of the plant: process, supplies, maintenance, instruments, quality. None of them can step outside what you sign.

  1. Watch

    The agents read what the plant already measures and check that each instrument can be trusted before using it.

  2. Act within limits

    Small, low-risk adjustments happen on their own, inside a signed range, step and minimum interval.

  3. Ask you

    Anything bigger waits for a person. You see what, why and what it costs, and you approve or decline.

  4. Record

    Every proposal, action and decision goes into a log that shows who, when and exactly what was written.

One tap is a decision, not autopilot

Between the agents and the plant there is a guardian. It is the only part allowed to write, and it checks every request against the limits you signed.

Emergency stops and safety systems are off limits. Always.

  • You sign the limits

    Which valves and pumps, how far, how fast and how often. Outside that, the answer is no.

  • Bad sensor, no action

    If an instrument is frozen, out of range or flagged as bad, anything that depends on it is blocked.

  • The numbers are calculated

    Figures come from process calculations. The language model only explains them and its text is rejected if it changes one.

  • Your data stays yours

    The models can run on a computer inside the plant. Client data is not sent to free or public AI services.

  • Shadow mode first

    Before it touches anything, Lux only records what it would have done, so you can compare it with your operators.

  • Attacked on purpose

    We test the guardian with adversarial cases, and each one that ever worked is now an automatic test.

Where we are today

We would rather show the real stage than promise more than we have tested.

Working now

On a digital twin, with synthetic data.

  • Agents that propose and act on a simulated plant
  • Guardian with signed limits and approvals
  • Instrument health checks on the server
  • Shadow mode and tamper-evident log
  • Our own language model, running locally
  • Automated tests, including security attacks

Building next

The road to version 1.0.

  • Reading a real plant over OPC UA and Modbus
  • Writing to one low-risk loop, with read-back and automatic rollback
  • Heartbeat: if Lux goes quiet, the PLC and the operator keep control
  • Secure approval from a phone, away from the plant
  • Shadow-mode report: Lux compared with the operator

What we are looking for

To get from prototype to first plant.

  • A pilot plant in Central America
  • Cloud infrastructure for secure remote approval and multi-plant monitoring
  • Industrial cybersecurity review before writing to a real plant

First case: a water plant at a mine

A reverse-osmosis plant that delivers 8 cubic metres of water an hour to the mine's recovery process. If the water stops, the process stops. That is the plant our digital twin reproduces.

  • Mining
  • Water
  • Sugar and cogeneration
  • Food and beverages
  • Energy
  1. Raw waterTank level
  2. PretreatmentCartridge filters: maintenance agent
  3. Chemical dosingpH and antiscalant: process and supplies agents
  4. RO membranesReject valve and cleaning: process agent, with your approval
  5. Permeate qualityConductivity: instrument agent
  6. Water to the mineReserve and warnings: quality agent

Who we are

Lux Cognition is a Central American company based in Managua. We put process engineering in the field together with artificial intelligence.

Oscar Efrén Narváez

Co-founder · Director and president

Leads strategy, product development and market research.

efren@luxcognition.com

Tell us about your plant

If you own or run a plant and would like your time back, or you want to follow what we are building, write to us.

General contact

join@luxcognition.com

Managua, Nicaragua · We answer in Spanish and English.