John George — Co-Founder

Currently at capacity and not taking new clients. Follow my work here or get in touch for future availability. --- I build production voice AI systems on Pipecat and LiveKit. Full SaaS products, custom infrastructure, and migrations off managed platforms like Vapi and Retell. The problem is that building voice AI on open-source frameworks is genuinely difficult to get right. You're figuring out which STT/LLM/TTS combinations work, where to place your servers for optimal latency, debugging why turn detection keeps interrupting mid-sentence, and learning what actually works at scale versus what looks good in demos. I've been doing this for nearly two years. I've debugged the failures, optimised the pipelines, and built the boilerplate. I also write about AI-assisted development and software security. If you've found me through one of those posts, that's the same engineering mindset applied to different problems. Tech stack: Python, Next.js, TypeScript, Supabase, Vercel, with Pipecat or LiveKit depending on your needs. I offer consultancy for architecture and migration planning, or implementation where I build it and hand it over. Sometimes both. If you're building voice AI and want to talk through your approach, get in touch.

Stackforce AI infers this person is a SaaS-focused Voice AI developer with expertise in software architecture and emerging markets.

Location: Folkestone, England, United Kingdom

Experience: 19 yrs 1 mo