Skip to content

EMARKSS LTD · UK company · Est. 2025

EMARKSS builds Lumnya, the AI workforce for online stores.

We are a UK startup with a single focus: Lumnya, an AI-native team that sells, supports and runs the back office of Shopify stores, with the owner in control of every action that matters.

Product
Lumnya, private beta
Built with
Claude Code

01 · The company

The company

EMARKSS is a UK company with one product and no distractions. Everything we build goes into Lumnya, an AI workforce that acts inside the store, remembers what it learns, and explains every move to the owner.

02 · The product

Lumnya. Your store's digital workforce.

A customer-facing AI that sells and supports shoppers around the clock, and a back office of specialist agents. Anything that touches money or customers waits for the owner's approval, and every action is logged.

lumnya · storefront + approvals
Illustrative interface: a shopper asks the storefront assistant for a refund on a damaged order; Lumnya prepares the refund and holds it until the store owner approves, then records the decision in the journal.

03 · The workforce

A back office that never clocks off.

Specialist agents scan the store on a schedule. They forecast, analyse and recommend, each with its reasoning attached, share what they learn with one another, and act only when the owner says yes.

Agent activity Snapshot · per store
Illustrative interface: agents report a reorder awaiting approval, an explained revenue change, an unmet demand signal, a clearance plan awaiting approval and a drafted FAQ.
  • Live

    Inventory Forecaster

    Watches stock across the catalog and proposes priority-ranked reorders, with the reasoning behind every number.

  • Live

    Profit Lens

    Maps the true margin of every product, from the costs the owner enters.

  • Live

    Liquidation Planner

    Finds dead stock and drafts bounded clearance plans for approval.

  • Live

    Bundle Discovery

    Finds products that really sell together, from actual order pairs.

  • Live

    Demand Radar

    Hears demand in storefront conversations that the catalog does not cover.

  • Live

    Revenue Detective

    Explains every revenue change by its drivers: volume, basket, refunds and mix.

  • Live

    Self-Learning FAQ

    Turns questions the assistant could not answer into FAQ drafts, one click to approve.

  • Next
    • Pricing Optimizer
    • Churn-Risk Scorer
    • Sentiment Pulse
    • Anomaly Detector
    • Executive Summary

Plus the manager console

  • Ask in plain language
  • Goals set in words
  • Give it a file, not a task
  • Hourly store watchdog
  • Weekly report that suggests changes

04 · How we build

Built with Claude Code. Runs on the Claude API.

Lumnya has been built over several months with Claude Code, Anthropic's agentic coding tool, which our team uses to design, write, test and ship the product. The intelligence inside Lumnya runs on the Claude API.

What Claude does inside Lumnya

  1. 01

    Conversations with shoppers

    Selling, support, orders and returns on the storefront, in the shopper's language.

  2. 02

    Agent reasoning

    The analysis and the "why" behind every recommendation an agent makes.

  3. 03

    Tool use behind confirm gates

    Actions in the store are proposed as exact changes and wait for the owner.

  4. 04

    Solving the merchant's problems

    Tracing an issue in the business to its cause and proposing the fix.

  5. 05

    Recognising customer patterns

    Reading behaviour and conversations to see what shoppers want.

Multi-model engine

Claude API

Each task is routed to the Claude model that fits it. Routine work goes to fast models; high-stakes steps such as checkout, cancellations and returns go to the stronger ones.
Development
Built with Claude Code
How the product is engineered.
Intelligence
Runs on the Claude API
Sonnet 5.5, Opus 5.5 and Fable 5.1, routed by task.

05 · How it works

Three steps to a workforce that answers to you.

  1. 01

    Install

    Add Lumnya to a Shopify store and connect it. No code, and nothing is written into the theme files.

  2. 02

    Learn the store

    Lumnya reads what Shopify already knows. The owner adds what it cannot: hours, couriers, return window, discount limits. Memory builds week over week.

  3. 03

    Work with approval

    Agents recommend with their reasoning. The owner reviews the exact change and approves. Lumnya acts, logs it and remembers the outcome.

06 · Progress

One product, shipped month by month.

From incorporation to a private beta with a full agent workforce. The public roadmap and changelog live on lumnya.ai (opens in a new tab).

  1. EMARKSS LTD is incorporated

    Registered at Companies House as company 16181893. View record on Companies House (opens in a new tab)

  2. Customer AI goes live

    Selling and support on the storefront, connected deeply to Shopify.

    • Customer AI
    • Shopify deep integration
  3. The owner gets the controls

    Every consequential action previews the exact change and waits for a yes.

    • Confirm-gated actions
    • English, Greek & German
  4. Agents start to remember

    The first back-office specialist, and a per-store memory for every agent.

    • Inventory Forecaster
    • Agent Memory
  5. A full back office

    Margins, dead stock, autonomy on a dial, and a manager that takes goals in words.

    • Profit Lens
    • Liquidation Planner
    • Autonomy dial, journal & undo
    • Trust panel
    • Store Watchdog
    • Goals set in words
    • Give it a file
    • Standing instructions
  6. Sharper and safer

    The weekly report now suggests what to change. Security strengthened after an independent review.

    • Weekly report with suggested changes
    • Platform security
    • Smarter chat & FAQs
  7. Now

    Private beta, waitlist open

    Next on the roadmap: Pricing Optimizer, Churn-Risk Scorer, Sentiment Pulse, Anomaly Detector and Executive Summary. On the ballot: WooCommerce and a multi-store view.

07 · Principles

What Lumnya is built to do.

  • 01

    Acts.

    Does the work inside the store, instead of reporting problems back to the owner.

  • 02

    Remembers.

    Keeps a private, per-store memory that grows more useful week over week.

  • 03

    Explains.

    Every recommendation carries the data and the reasoning behind it.

  • 04

    Owner in control.

    Money and customer actions wait for approval. Autonomy is a dial the owner sets.

08 · Team

The people building it.

Private beta · Shopify first · EN · EL · DE

Meet Lumnya.

See the product, or join the waitlist for an invite when the next wave of stores opens.