← All work

SpaceBrain — CRM, Marketing & AI in One Platform

A venture I co-founded: 50+ business tools in one subscription, from AI CRM, funnels and email to agreements, payments and AI receptionists that answer real calls 24/7. Live with customers in North America and Thailand.

PythonNext.jsTypeScriptVoice AIRAGStripe

Overview

SpaceBrain is a product I co-founded: 50+ business tools in one subscription, built around one shared customer record. CRM, forms, funnels, email, calendar, agreements, payments, courses and social media sit in the same workspace, together with an AI receptionist that answers the phone around the clock, takes bookings, answers questions from a knowledge base and hands off to humans when it should.

It's not a demo. It answers live calls for real businesses in Canada, the USA and Thailand, and there's a free-forever plan at spacebrain.ai. I wrote up the problem it solves and why we built it.

What it does

  • Voice agents on real phone lines — low-latency speech-to-speech conversations over telephony, with barge-in handling and graceful fallbacks.
  • RAG knowledge bases — businesses upload their docs and the agent answers from them, with grounding rather than guessing.
  • Bookings & actions — the agent doesn't just chat; it books appointments and captures structured leads.
  • The rest of the business in the same record — forms, funnels, email marketing, automations, agreements, payments, memberships, courses and reputation management, plus white-label delivery for agencies.
  • Full SaaS layer — onboarding, Stripe billing, usage metering, affiliate tracking, an admin dashboard, and an embeddable web chat widget.

You can try that widget on this site: the chat assistant in the corner of every page runs on SpaceBrain, answering questions about my work from a knowledge base.

My role

Co-founder and lead engineer. I build and run the technical system: the voice pipeline, a Python backend of nearly a hundred domain modules covered by more than 27,000 automated tests, the Next.js dashboard, billing webhooks, deployment, and observability. Running a 24/7 voice product teaches you very quickly where distributed systems actually fail — webhook races, queue backlogs, replay attacks, latency budgets measured in milliseconds.

Why it matters for your project

If you're considering an AI product, I've already paid the tuition: prompt design, RAG quality, latency engineering, cost control, and the operational reality of LLMs in production.

Next step

Have a project that needs to actually ship?

Tell me what you're building. I'll reply within one working day with an honest take on scope, timeline, and whether I'm the right person for it.