Catalog or content app
Information lookup, no transactions or complex accounts.
iOS and Android through publication in both stores. The last one we built and maintained has more than two million active users across ten countries.
2M+ active users on the last one we built · 4 apps published on the App Store and Google Play · 10 countries in simultaneous operation
Built on
Ninety seconds: who we are, how we work and what you get at the end.
An email, an order or an alert comes in. The agent queries your systems, decides by your rules, asks for approval when needed and executes. Pick a case and watch it run.
Example: a customer asks what their policy covers. The system answers with the exact page of the contract and opens the case.
context
product = motor_premium
84 pp → 612 chunks
Asked for approval · human
log · 5 steps · who and when
Swipe to follow the flow →
01
Apple rejects for reasons that are on no obvious checklist, and each rejection costs days. You need to know what they check and prepare the submission for it.
02
Apple charges USD 99 a year, Google USD 25 once. They must be in your company’s name: if they belong to whoever built the app, the app is not entirely yours.
03
Every year iOS and Android change things that break apps. An unmaintained app stops working on its own within about two years.
04
Losing the Android signing key means never being able to update the app again. They are handed to your company and documented.
01
How we work
React Native for most cases: a single codebase for iOS and Android, so the team maintains one app instead of two. Native only when the case genuinely demands it — heavy graphics, intensive sensor use, very specific system integrations — and we justify it before proposing it.
And mobile almost never comes alone: it needs backend, authentication, push notifications, payments or integration with what you already have. We build that too, so you are not coordinating two vendors who blame each other when something breaks.
We work from Medellín with companies across Colombia, and remotely with teams in the United States and the rest of Latin America. Publishing to the App Store and Google Play happens under the company’s own accounts, wherever it is based.
Price is driven by three things: how many features it has, how many systems it connects to, and whether it needs its own backend. Here are the scopes and their timelines; the range for your case comes out of the free 15-minute call.
Information lookup, no transactions or complex accounts.
User accounts, payments, notifications, own backend.
App, admin dashboard and integrations with existing systems.
If your idea is not yet validated with real users, an app is a very expensive way to find out.
In many cases it is better to start with a mobile web experience or even a WhatsApp flow, confirm people use it, and build the app once you know which features matter.
The system around the model, not just the model.
MCP servers against the systems already in operation.
With the evaluation set run before it ever reaches production.
Architecture, evaluation criteria and cost per query before writing code.
Key takeaways
Building the app is half the work: the other half is store review, developer accounts, signing certificates and annual maintenance.
An unmaintained app stops working on its own within about two years, because iOS and Android ship yearly versions that break things.
Quarl uses React Native for most cases — one codebase for iOS and Android — and native only when the case genuinely demands it.
If the idea is not yet validated with real users, an app is a very expensive way to find out: a mobile web experience or a WhatsApp flow costs a fraction.
An AI system is worth what its sources are worth. These are the standard connectors; anything with an API or a database connects the same way, and what has no API is handled by file.
SAP
Enterprise ERP
Oracle
ERP and database
NetSuite
Cloud ERP
Salesforce
CRM and service
HubSpot
CRM and marketing
PostgreSQL
Database and pgvector
Microsoft SQL
Database
Snowflake
Data warehouse
BigQuery
Google data warehouse
Databricks
Data platform
Redshift
AWS data warehouse
Synapse
Azure data warehouse
Supabase
Managed Postgres
Workday
Payroll and HR
QuickBooks
Accounting
Sage
Accounting and ERP
Xero
Cloud accounting
Shopify
Catalogue and orders
WooCommerce
Catalogue and orders
Magento
Catalogue and orders
Stripe
Payments and subscriptions
Google Drive
Documents and folders
CSV y Excel
Flat files
Nothing in this category
We built and operated the assistant for a loyalty platform serving more than two million active users across ten countries.
We built the full pipeline: document ingestion and normalization, chunking, embedding generation, vector store on Azure AI Search and Pinecone, and retrieval with grounded generation on LangChain.
We held it above 99% availability for three years.
RAG systems →React Native for most cases, because a single codebase for iOS and Android is the difference between one project and two, in cost and in maintenance. Native when the case demands it: heavy graphics, intensive sensor use, very specific OS integrations or extreme performance requirements. We justify it with concrete reasons before proposing it.
Build time depends on scope, six weeks to six months. App Store review usually takes one to three days, but a rejection adds days and sometimes weeks if it forces product changes. Google Play is faster. We prepare submissions for Apple’s review criteria, which is where most projects get stuck.
Your company’s, always. It seems like a minor detail and it is not: if the accounts belong to the vendor, you cannot change vendors without losing the app and its reviews. We set the accounts up in your name from the start and hand over documented access and certificates.
Every year iOS and Android ship versions that break things, and the stores raise their minimum requirements. An unmaintained app stops working on its own within about two years. So maintenance is not optional; who does it is. Your internal team can carry it with the documentation and access we hand over, or you can leave it with us under contract. Annual cost usually runs 15% to 20% of build cost, and it is the one service of ours we recommend not skipping.
Yes. An app almost never lives alone: it needs authentication, an API, a database, push notifications and sometimes payments. We build it end to end so you are not coordinating two vendors who blame each other when something breaks.
Yes, and it is one of the highest-value combinations we see: an in-app assistant with the user’s context and permissions, semantic search over your content, or data extraction from the camera. We apply the same measurement discipline as in any AI project.
We will suggest not building the app yet. Validating with a mobile web experience or even a WhatsApp flow costs a fraction and tells you which features actually matter. Building the app afterwards, knowing that, is cheaper than building it twice.
It gets fixed, it gets built, or it is not worth it. And if the diagnostic does not reach three actionable findings, it is not charged.
We use cookies to improve the user experience. Privacy