Our stack

The stack behind our agents.

The models, tools and infrastructure we actually use to build AI agents and products — what each one does for us and where it runs. No affiliate links, no sponsorship: when a tool stops earning its place, it goes.

Principles

Models
Chosen per task, never one provider for everything
Hosting
Containers we run, not serverless platforms
Data
Open-source databases we can dump and restore
Messaging
Official WhatsApp Business Platform for clients
AI models

Models and speech

ToolWhat it does for usWhere we use it
Anthropic ClaudeReasoning and tool use for agents; our team's coding assistantRestaurant booking assistant, Mama Bot, the gym agent, the training platform's assistant, our own engineering
OpenAI modelsStructured extraction and generationMenu import and writing in Vimus, agents where it fits the task
Open-weight models via OpenRouterFast, low-cost models for high-volume internal assistantsThe assistants of Takat OS
Whisper (open source)Speech-to-text on a machine we runThe meeting notetaker, with no audio sent outside
ElevenLabsNatural voice for a spoken agentThe voice-ordering prototype on WhatsApp calls
Google text-to-speechVoice replies to voice notesThe gym agent's spoken answers
Conversations

Messaging and email

ToolWhat it does for usWhere we use it
WhatsApp Business Platform (Cloud API)The official way for a business to talk to customers on WhatsAppClient agents: bookings, ordering, gyms, insurance follow-ups — see WhatsApp AI agents
MattermostOpen-source team chat we host, with our bots insideTakat's own team chat and its assistants
Resend, SendGridTransactional emailBooking confirmations, notifications, forms
Infrastructure

Where it runs

ToolWhat it does for usWhere we use it
Docker ComposeOne stack per project: app, database, workersEvery production project
CaddyThe single web gateway, with automatic TLSIn front of every site and app
GitHub ActionsBuilds and tests every change, tags images with the commitEvery repository; nothing is built on a production server
Google Cloud (virtual machines)The servers our stacks run onProduction, previews and development environments
CloudflareDNSOur domains and those of the products we run
PostgreSQL, MySQL, RedisDatabases and cache, each in its project's own containerBooking platforms, menus, CRMs, bots
NocoDBA database with an API and an admin viewThe data behind Takat OS
SentryError trackingClient platforms
Applications

What we build with

ToolWhat it does for usWhere we use it
React, TypeScriptInterfaces for staff, owners and customersBooking dashboards, Vimus, this website
Next.jsSites and apps that must be fast and indexableAssermento, the creator platform, multi-brand sites
ViteFast single-page apps and prerendered sitesVimus, this website
AstroStatic e-commerce and content sitesA custom online shop
Stripe, PaystackOnline payments, including African cards and walletsA custom online shop, Vimus subscriptions, hospitality platforms
How we choose

Fewer moving parts, data we can restore

We prefer containers we run to serverless platforms: one place to look when something breaks, one bill, and data we can dump, restore and move. We pick a model per task — the best reasoning where the agent must act, a fast and cheap one where it must sort, an open-source one where the data must not leave. And every tool has to be replaceable: an agent's logic lives in our code, not inside a vendor's builder.

How these pieces fit together is described on how we build.

FAQ

Frequently asked questions

  • Which AI model does Takat use?

    Several, chosen per task: Anthropic Claude for agents that reason and act, OpenAI models for extraction and generation where they fit, open-weight models for high-volume internal assistants, and open-source Whisper for transcription on our own machine.

  • Is Takat tied to one AI provider?

    No. The agent's logic, tools and data live in our code and your database, so the model behind a task can be changed.

  • Do you use no-code chatbot builders?

    No. Our agents are written as code, tested, versioned in git and deployed as containers, so they can be reviewed, rolled back and moved.