AI agents & assistants
Agents that answer, sell, book and triage over WhatsApp, web chat and your own tools.
AI-native software studio
We build our own products.
Then we build yours.
Farcholabs designs, builds and ships mobile apps, web platforms and AI agents, from a written spec to a release you own: the code, the accounts and the docs.
AI triage of new leads with Claude
- donespecPLAN.md §8
- doneplanprompt triage-v1
- donechanges27 files · +1,619 −27
- donetests15 new tests · CI green
- queueddeployoff until evals pass
Agents wrote the change. An engineer reviewed it and merged it.
A replay of pull request #5 in this site's own repository, September 2026.
What we build
Apps, platforms, agents and automation, built one way.
01 02 Mobile apps
iOS and Android apps from one codebase, with the API and admin behind them.
03 Web platforms & SaaS
Multi-tenant platforms, portals, dashboards and admins, with auth, roles and billing.
04 Automation & data
Workflows in n8n, Make or Python between the tools you use, and numbers that add up to the cent.
How we build with AI
AI drafts. An engineer decides.
Agents do the typing. An engineer writes the spec, reviews every change and answers for what ships.
- 01
Frame
The spec comes before the code.
We write down the goal, the users, the scope, the risks and the decisions behind them, and turn it into a plan with milestones. You read it and approve it before anything is built.
- 02
Build in loops
Agents write the code. An engineer owns every change.
AI agents work through the spec in short loops. An engineer plans each change, reviews every diff and decides what merges. You follow along on a preview build as the work lands.
- 03
Prove
Nothing merges on a red check.
Tests, type checks and accessibility checks run on every change. AI features also have to pass their evals before they are turned on, and again whenever the prompt or the model changes.
- 04
Ship & hand over
You get a product you can run without us.
We deploy, watch it in production and hand over the repository with its docs and the agent notes (CLAUDE.md) that let any engineer, or any agent, keep building on it.
Guardrails on every project
- Written specs
- Architecture decision records
- Tests in CI
- Evals for every AI feature
- A person reviews every change
Engagements
Choose how we work together.
Every engagement starts with a written proposal: scope, milestones and a fixed price for each one.
Product Sprint
Turn an idea into a spec, a prototype and a priced plan.
From $5,000
- Written spec and decision records
- Clickable prototype or technical spike
- Phased plan with a price
Fixed price, paid up front. Credited in full to a Build that starts within 60 days.
Build
A product built phase by phase, from first release to v1.
From $12,000
- Fixed scope per milestone
- Tests, CI and docs
- The repository and its IP
A deposit, then billed by milestone.
Retainer
Ongoing engineering for a product that is already running.
From $3,500 a month
- Features, fixes and upgrades
- Model and prompt updates, re-run against the evals
- Monitoring in production
Billed monthly, in advance.
Essential Agent
An agent that answers your customers on one channel, hosted and looked after by us.
From $1,500 + $250 a month
- One agent on one channel: WhatsApp or a chat on your website
- Answers from your catalog, FAQ and documents, and hand-off to a person
- Evals it has to pass before it goes live
- Hosting, monitoring and up to 2 hours of changes a month
Setup paid up front, then billed monthly in advance. A 3-month minimum, then cancel any month.
AI Agent
One agent for one job, wired into your systems and built in your own accounts.
From $8,000
- Agent with tools and retrieval
- Eval suite in CI
- Review console and hand-off to a person
A deposit, then billed by milestone.
Essential Automation
Up to three workflows between the tools you already use, watched and kept running by us.
From $1,000 + $150 a month
- Up to three workflows in n8n or Make, in your own account
- Retries, alerts and docs for each one
- Fixes when one of your tools changes its API
- Monitoring and up to 1 hour of changes a month
Setup paid up front, then billed monthly in advance. Cancel any month.
Guides
What it costs and how we build it.
Prices from our own list, how we build and test, and what to ask before you hire anyone.
Guide 10 · 2 min React Native or separate native apps for a first release.
For most first releases we build one React Native app for both stores. This guide explains why, what you give up, and the cases where two native apps are the better call.
Guide 09 · 2 min How to keep your app's billing in step with Stripe.
Stripe decides what was paid. Your app keeps a copy, and the copy drifts unless something keeps it honest. This is how the billing module in this site's own code does it, and how to check payouts against the payments inside them.
Guide 08 · 3 min A build log, explained: how AI agents and an engineer ship a pull request.
The home page of this site replays a pull request: spec, plan, changes, tests, deploy. This guide walks through that pull request, the one that added AI triage of new leads, and what a person did at each step while agents wrote the code.
FAQ
Before you ask.
Who owns the code?
You do. The repository, the code and the IP we write for you are yours once each milestone is paid, as the proposal and our terms set out. We keep only general know-how and the tools we had before. An Essential Agent is the exception: it runs on our platform, and what's yours is your content and your customers' conversations, which we export for you whenever you ask.
Do you sign NDAs?
Yes. We sign yours, or send ours, before you share anything confidential.
How fast can you ship?
It depends on the scope, so we don't promise a number before we've read the brief. The proposal gives each phase a range of weeks, and the first milestone is always something you can use.
How much does it cost?
Each engagement has a starting price on the pricing page. After a scoping call, the proposal fixes the price of every milestone before any work starts.
What if the AI gets it wrong?
It does, sometimes. That's why an engineer reviews every change before it merges, tests run on each one, and AI features in your product ship with evals, fallbacks and a way to hand off to a person.
Which time zone do you work in?
We're based in Medellín, Antioquia, Colombia, on UTC−5 all year (Colombia has no daylight saving).
Which models do you use?
Claude, from Anthropic, both to write code and inside the products we build. Where a product needs it, a second provider takes over when the first one fails. We pick per task and measure with evals.
Have something to build? Tell us about it.
A person reads every brief and replies with questions or a first plan.