palmoraproperty.com
Palmora Property · Real estate
A portal that speaks to buyers from Europe, Russia and China, with no full-time translator
÷ 60 translation cost, by switching to the right AI model
Read the case study
Web agency · Automation · AI
AI that works for your people, for a business that grows without growing its costs.
Get the full benefit of AI in your industry. Websites, automations and AI agents for teams of 20 to 500 people: machines take the repetitive work, your people keep what matters.
- Prices announced upfront
- Delivered in weeks
- People stay in charge
Lower operating costs, with AI where it removes work.
We don’t sell technology: we remove hours of data entry, redundant tools and manual follow-ups. Every project is measured by what it makes disappear.
Sound familiar?
"We pay for tools that don’t talk to each other, and re-key everything by hand."
"Quotes, invoices and follow-ups eat whole days."
"The website brings nothing in, the catalogue is a mess, and nobody has the time."
Websites and e-commerce
What changes A site that updates itself, in every language, with no translation agency for each edit.
Brochure sites, content sites, online stores.
- · Full redesign
- · Catalogue and product pages
- · Checkout flow
Process automation
What changes Re-keying disappears, and every workflow shows what it costs to run.
Connect your tools and remove manual data entry.
- · Task mapping
- · Workflows in production
- · Alerts and digests
CRM and ERP
What changes No per-user licence that climbs with headcount; the tool grows without the bill following.
Management tools deployed without per-user licensing.
- · Twenty, Dolibarr, Odoo
- · Data migration
- · Team training
AI agents and bots
What changes Follow-ups, data entry and routine answers run on their own, reviewed by a human where it matters.
Agents that take a task from end to end.
- · Assistant on your documents
- · Telegram and WhatsApp bots
- · Document extraction
Migration to a static site
What it changes No more security updates to chase or server to watch, and pages that display fast everywhere, including from China.
Leave WordPress or Drupal for an Astro site: plain HTML, addresses kept, zero plugin to patch.
- · Migration with unchanged addresses
- · Astro + Sanity headless
- · Publishing chain with rollback
SEO, audit and AI visibility
What it changes Every piece of content leaves with its source and its target page, and the month’s work comes out of your Search Console data, not an agency calendar.
A site that brings in requests instead of costing a subscription.
- · SEO audit with action plan
- · Continuous content engine
- · Citations in ChatGPT and Perplexity
Web and mobile apps
What it changes Your first users test a real app before you invest in the stores.
The tool off-the-shelf software won’t give you, on everyone’s phone.
- · One codebase: browser, iPhone, Android
- · Installable web version within the first weeks
- · Publishing on the App Store and Google Play
Re-keying
What does copying data cost you?
Set the three sliders with your own figures: the amount is the time spent each year copying from one tool to another.
a year spent copying, roughly.
Estimate based on 46 working weeks a year. It is not a savings promise: it is the order of magnitude an automation can take back.
÷ 60
on our translation costs
Measured on the Palmora Property multilingual pipeline (6 languages) after switching models, with 95% of segments served from cache.
Source: Brain Plus AI, measured on the Palmora Property translation pipeline
AI at the right cost.
For every task, the most economical AI model that meets the quality bar. Measured, capped, reversible.
-50 to -90% in provider discounts applied by default
Batch processing, caching and off-peak windows, at the providers' official rates. Our automations are built to use them out of the box. Source: public pricing from model providers.
1 meter with every automation we ship
Cost per run and per month, an agreed cap, an alert before it is reached. You see what each system costs, continuously. Source: Brain Plus AI, every automation delivered.
An agent takes a task from end to end.
Not a chatbot that answers questions. An agent that receives, processes, records and confirms, with no human involvement in the repetitive part. We don’t put AI everywhere: only where it removes work with no added value.
- Email received , then Sorted by Jev
- Sorted by Jev (quote), then Quote in the CRM
- Sorted by Jev (complaint), then Draft reply
- Sorted by Jev (unsure), then A person decides
- Quote in the CRM , then Sales rep alerted
- Draft reply , then Reviewed, then sent
A request arrives by email. Jev files it, a rule creates the quote in the CRM, a language model drafts the reply to a complaint, and a person decides as soon as the model hesitates.
- Site form , then Record in the CRM
- Record in the CRM , then Ideal customer?
- Ideal customer? (priority), then Sales rep alerted
- Ideal customer? (unsure), then A person decides
- Ideal customer? (to nurture), then Email sequence
A contact fills in the site form. A rule creates their record, Jev compares them with your ideal customer, and the sales rep is only called for priority contacts; the others get an email sequence.
- Invoice received , then PDF read
- PDF read , then Matching
- Matching (matches), then Accounting entry
- Matching (mismatch), then Mismatch reviewed
A supplier invoice arrives. A language model reads it, the code compares it with the purchase order, and only a mismatch needs a person to look.
- Order confirmed , then Invoice created
- Invoice created , then Payment received?
- Payment received? (yes), then Shipping started
- Payment received? (no), then Customer reminded
An order is confirmed. Invoice, payment check, shipping or reminder: everything follows rules, with no call to a model. Many processes look like this one.
- Free-form request , then AI agent
- AI agent , then Safe action?
- Safe action? (safe), then Action carried out
- Safe action? (risky), then Human approval
- AI agent : Model
- AI agent : Memory
- AI agent : Tools
A request written in free form. The agent searches over several rounds, with its model, its memory and its tools, but every action goes through Jev before it leaves, and anything risky waits for a person.
Swipe to follow the diagram
- Rule, no AI
- Decision model
- Language model or agent
- Person
Plugged into your tools, not the other way round
Twenty Odoo Dolibarr HubSpot Shopify Stripe Gmail Outlook Google Sheets Excel Notion Airtable
Slack Microsoft Teams Telegram WhatsApp PostgreSQL Supabase Jev Clef Mistral Claude Qwen Webhooks et API
Telegram bots and virtual employees
Hire virtual employees for your time-consuming tasks.
We build Telegram bots that work like employees: they take a request, do the repetitive work in your tools and only ask you to approve. Data entry, reminders, filing, reports: what takes you hours every week happens without you. Here is a bookkeeping employee we run every day.
Accounting Agent bot
Today
09:42 
Receipt analysed. Here's what I extracted:
- Vendor
- Petrol station
- Amount
- €68.40
- Date
- 07/10/2026
- Category
- Fuel
✓ Approve 09:43
Expense booked proof filed · ledger updated 09:43
What the bookkeeping employee does, step by step
Telegram you
Photo of the receipt or invoice
AI agent automatic
Extraction, analysis and proposal
Telegram you
Validation or correction
Google Drive automatic
Photo saved to the right cloud folder
Odoo automatic
Data recorded in the accounting software
Google Sheets automatic
Raw data saved to the shared files
Gmail automatic
Documents emailed to the accountant
Illustrative example, made-up amounts. Each service can be swapped for the one you already use: Pennylane, Dolibarr, Slack, Notion…
See the Telegram and WhatsApp bots offerProjects delivered, not promises.
Each study documents the real context, the work done and the measured result. No extrapolated figures.
See all case studies peluche-kingdom.com
Peluche Kingdom · E-commerce
A taken-over store that sells again, now open to Germany and Scandinavia
+137% sales in September 2026, against the March to August average
Read the case study
challengeus.app
Challenge Us · App
A fast-growing app, with ever more users who keep coming back
Group challenges logged and followed live, a food diary with barcode scanning, a weight curve and routines to keep together. No ads and no public profile, on the web, with iOS and Android on the way.
Read the case study
How it works
An audit first, then two-week sprints.
Not sure what pays off most?
The digital and AI maturity audit maps your processes and ranks the projects by impact, with an honest GO/NO-GO.
The Audit Sprint in detail-
Scoping
A 20-minute diagnostic, then the Audit Sprint: your processes mapped, three priorities, measurable goals and a go/no-go.
-
Design
The architecture validated with you before we build: only the modules and workflows you need.
-
Sprints
Two weeks per sprint, something that works at the end of each, training alongside.
Results
Assisted go-live, then check-ins at 30 and 90 days against the goals set at scoping.
Fixed scope
What we agree on is what gets delivered. No scope creep mid-project.
Short sprints, precise deliverables
Every sprint produces something that works. No months with nothing to show.
No technical lock-in
You keep ownership of your tools, your data and your access. At all times.
Prices fixed before we start.
All offers, scope and pricesBlog: guides and field notes, sources included
20-minute diagnostic
Twenty minutes on a call, two or three concrete leads, no strings attached.
We look at your situation, identify what can be automated or improved, and tell you plainly whether we can help.
No endless form: a few fields are enough.