Our Tech Stack and Why We Picked It
The tools we use to build automations, dashboards, and AI systems - and the reasoning behind each choice.
We get asked about our technology choices constantly. Here's a transparent look at what we use to build automations, dashboards, and AI systems-and why we picked each tool.
Automation: n8n, Make, and custom code
For most automation projects, n8n is our default. It's self-hostable, has excellent node coverage, and handles complex logic well. We use Make for simpler flows where clients need to self-manage. Custom code (Python, Node.js) when nothing else fits.
AI: OpenAI, Anthropic, and local models
We're model-agnostic. GPT-4 handles most conversational AI. Claude excels at document analysis. For cost-sensitive applications, we deploy smaller open-source models. The choice depends on the use case, not vendor loyalty.
Frontend: Next.js and React
When we build dashboards or portals, it's Next.js 90% of the time. Great developer experience, excellent performance, and easy deployment to Vercel. React component libraries speed up development.
Database: Supabase and PostgreSQL
Supabase gives us PostgreSQL with a great dashboard, auth, and real-time features built in. For simpler needs, Airtable. For complex enterprise requirements, dedicated PostgreSQL instances.
Key Takeaways
- Tools should match the problem, not the other way around
- Self-hostable options give clients more control
- Avoid vendor lock-in where possible
- Boring technology is often the right choice
Curious whether AI makes sense for your business?
Book a 30-minute call. We'll talk through what you're dealing with and whether there's something worth building.
Book a Discovery Call →