CLIENT WORK / EXARABUILT IN DUBAI. BUILT AROUND YOU.

FROM EVERYDAY FRICTION TO WORKING SOFTWARE

A better
working day.

The stock sheet. The endless follow-up. The report that eats every Monday. We build the software that gives your team that time back.

see the work tell us what’s slowing you down

THE SHAPE OF A BETTER DAY

REAL INPUTS → CLEAR RULES → USEFUL OUTPUTS

SELECTED CLIENTS

ANATIONZOpenSpaceFEASIBLE

how we work

01 / selected work

Real businesses.
Working systems.

01 / 03

ANATIONZ × EXARA

retail replenishment / delivered

Three weeks of manual allocation. Now: seconds.

Anationz distributes clothing stock across retail chains. Every refill cycle meant reading separate sales files, deciding what to send, and checking every line against warehouse stock by hand.

Clean retailer sales inputs into one known format.
Recommend refills per store from what is actually moving.
Cap every line to real warehouse stock before it leaves the system.
Generate the refill sheets and brand reports the team already uses.

cycle time: ~3 weeks -> seconds
retail coverage: 4+ chains
recommendations: 100% stock-capped

OPENSPACE × EXARA

property command center / delivered

Twenty five properties. One screen.

OpenSpace manages residential units across the UAE. Rent, repairs, cheques and tenant conversations lived in WhatsApp threads, notebooks and one person's memory.

Bring the full portfolio into one command center.
Track rent, repairs, cheques and tenant issues in one place.
Turn every operational task into an open loop that has to close.
Keep contractor history and UAE-specific workflow visible to the team.

portfolio: 25 units, one view
every job: open -> done
built for the UAE: Ejari · DEWA · cheques

FEASIBLE × EXARA

provable planning / delivered

An answer that comes with proof.

The rules a business must follow live in contracts, policies and regulations. Checking a plan against all of them by hand is slow, and nobody fully trusts the result.

Translate business rules into formal constraints a computer can check.
Return plans that satisfy every rule by construction.
Attach a signed proof to every answer.
Let anyone re-run the verification on their own machine.

verification: 168 tests pass
rule pack: VALID
plan: OPTIMAL

ANATIONZ × EXARA

Three weeks of manual allocation. Now: seconds.

01 / the problem

Each refill cycle, someone read separate sales files, worked out what was moving, decided what each store should get, and checked every line against what the warehouse could actually supply. It took about three weeks, every cycle. The hard part was not making a longer list. It was making sure every suggestion stayed possible when all the stores compete for the same finite stock.

02 / what we built

Clean every retailer file into one known structure instead of trusting messy inputs.
Work out what each store needs from the latest sales movement.
Cap every suggestion to what the warehouse really holds before output is generated.
Land the result as the team's own refill sheets and brand reports, not a new dashboard they have to remember to check.

03 / what changed

The recurring work moved from assembly to review. A cycle that took about three weeks now runs in seconds, and every recommendation is stock-capped by construction.

cycle time: ~3 weeks -> seconds
retail coverage: 4+ chains
recommendations: 100% stock-capped

04 / see for yourself

sample input.csv sample output.csv validation result

The samples are representative and redacted. Customer data stays private.

05 / inspect the decision

One refill decision, opened up. The structure mirrors the shipped system; the values are safe demonstration data, not client records.

inspect the system

One refill decision, opened up.

Representative redacted specimen. The structure and constraints mirror the shipped system; values shown here are safe demonstration data, not client records.

parsed successfully

download input.csv ↓

Representative sales and stock input
sku	store	sold / 14d	on hand	warehouse
SKU-101	Store 01	18	2	8
SKU-204	Store 01	12	1	3
SKU-330	Store 02	7	0	0
SKU-480	Store 02	4	5	10

Rows are normalized before the decision pass. Missing identifiers and malformed quantities are quarantined rather than guessed.

4 decisions emitted

download output.csv ↓

Representative refill output
sku	requested	send	reason
SKU-101	12	8	capped
SKU-204	8	3	capped
SKU-330	6	0	no stock
SKU-480	0	0	sufficient

The operational output is a refill sheet plus a report per brand, so the recommendation lands in the team’s existing workflow.

06 / what this case does not claim

This is a replenishment decision system, not autonomous purchasing.
Missing or malformed source data is surfaced, not filled with invented values.
The downloadable records are representative and redacted, not Anationz production data.
Final operational review stays with the client team.

got a version of this? tell us

OPENSPACE × EXARA

Twenty five properties. One screen.

01 / the problem

When something was disputed, there was nothing to point to. Who paid, what was promised, which repair was still open. It all lived in chats that scroll away and in one person's head. If that person was off for a week, the business forgot.

02 / what we built

Create one dashboard for the whole portfolio instead of scattered chats and notes.
Move every repair from open to done on a board the whole team can read.
Keep a contractor list that remembers who was quick, reliable and worth calling again.
Build around how the UAE actually operates: Ejari, DEWA and cheque-based reality.

03 / what changed

Twenty five units now sit on one screen. Every job has a state, an owner and a history with timestamps. Nothing depends on memory anymore.

portfolio: 25 units, one view
every job: open -> done
built for the UAE: Ejari · DEWA · cheques

04 / see for yourself

Client words coming here soon.

got a version of this? tell us

FEASIBLE × EXARA

An answer that comes with proof.

01 / the problem

A plan that breaks one clause in one contract can cost real money. Checking every plan against every rule by hand is slow, and the answer still comes down to trusting whoever checked it.

02 / what we built

Turn contracts, policies and regulations into a formal rule system.
Generate plans against that rule pack instead of reviewing them manually after the fact.
Ship every answer with a signed proof that can be checked independently.
Bind verification to the actual plan so the result is auditable, not just plausible.

03 / what changed

Trust stopped being a feeling. Anyone can re-run the check on their own machine and get the same answer. 168 tests pass, the rule pack is valid, and the plan is optimal.

verification: 168 tests pass
rule pack: VALID
plan: OPTIMAL

04 / see for yourself

VERIFY: OK. Signature valid, plan re-checked, rule pack bound.

got a version of this? tell us

does this sound familiar?

Your business runs on spreadsheets, WhatsApp, and one person’s memory.

01the stock sheet

You reorder from stock counts that are already wrong.

over-ordering · dead stock · stockoutshow we kill it

02the one person

One person holds half the business in their head.

single point of failurehow we kill it

03the group chat

Deals and approvals get buried in the group chat.

no proof · lost orders · disputeshow we kill it

04the month-end

Closing the month means days of chasing numbers.

three late nights, every monthhow we kill it

05the dead quote

Enquiries come in, then die waiting for follow-up.

paid leads going coldhow we kill it

06the monday report

The same report gets rebuilt from scratch every week.

hours lost, numbers nobody trustshow we kill it

the deal

A clear path from problem to production.

≤ 2 days

two working days to a straight answer

the promise

weeks

from yes to a running system, not quarters

the pace

your files

the first version runs on real data, not a demo

the proof

day rates. fixed scope, priced before we build

the pricing

a fixed price, agreed upfront

The scope and the number are set before we write a line of code. No day rates, no meter running, no surprise invoices.

you test before you commit

The first version runs on your actual files within days. You judge it on your own numbers, not on a demo with fake data.

delivered into your business

It plugs into what you already use, your sheets, your WhatsApp, your accounting. A running system in weeks, not quarters.

your code, your data

You own what we build. The code and the data are yours, nothing is locked behind us or rented back to you.

kept alive after launch

We monitor it, fix it and change it as your business changes. It still works next year, that is part of the deal.

Every case study here was delivered on these terms. Send the problem, get a straight answer within two working days.

tell us your problem

open problem line

What is taking too long,
costing too much, or
still being guessed?

Give us the unpolished version. We’ll read it, decide whether we can genuinely help, and reply by email.

A betterworking day.

Three weeks of manual allocation. Now: seconds.

Twenty five properties. One screen.

An answer that comes with proof.

You reorder from stock counts that are already wrong.

One person holds half the business in their head.

Deals and approvals get buried in the group chat.

Closing the month means days of chasing numbers.

Enquiries come in, then die waiting for follow-up.

The same report gets rebuilt from scratch every week.

a fixed price, agreed upfront

you test before you commit

delivered into your business

your code, your data

kept alive after launch

What is taking too long,costing too much, orstill being guessed?

Your problem is in our inbox.

A better
working day.

What is taking too long,
costing too much, or
still being guessed?