I help early- and growth-stage teams ship AI products, and the platforms they run on.

You built your MVP with Lovable, Bolt, v0, Replit or Cursor. It proved the idea, and now it has to hold up with real users, real money and a growing team.

It starts with a fixed-price audit and a ranked fix list in 48 hours. Then I either harden what you have in days or move it to a production codebase your team owns, and can stay on as your fractional CTO.

Track record: I run Next.js, Supabase and Vercel in production at BidSource and Yalla Bites, and built the Stripe billing and payouts at Instack AI and Yalla Bites.

See what's included and pricing

You need senior technical leadership, but not a full-time CTO yet.

I join your team part-time to own architecture decisions, review code, mentor engineers, and set up an AI-assisted workflow (Claude Code, Cursor, custom subagents) that your team can keep running.

Track record: Leading a four-engineer team at Instack AI today. At Coretava I led 6 to 10 engineers and split them into frontend and backend units: 50% fewer cross-team blockers and 25% faster sprints.

You want an assistant that can actually talk to your customers: in your app, on WhatsApp, or over the phone.

I build the whole path: LLM orchestration, real-time speech with LiveKit, Deepgram and ElevenLabs, telephony through Twilio and Asterisk, WhatsApp messaging, and the Stripe billing around it.

Track record: Cut Instack AI’s voice agent response time from 2 or 3 seconds to about one, and built DataQueue’s conversational AI platform end to end.

Deploys feel risky, the cloud bill keeps growing, or the database struggles under load.

Architecture reviews and hands-on fixes across Vercel, Supabase, AWS and GCP (GKE): Kubernetes, Terraform, CI/CD, message queues, and moving heavy queries off your main database.

Track record: At Coretava: 3K+ concurrent users on 40% fewer AWS resources, and analytics queries cut from 15 minutes to 30 seconds.