Not a demo, a product
There is a wide gap between an interesting LLM demo and a product people pay for. I work in that gap. Auth, billing, quotas, prompt versioning, guardrails, error states, and the unglamorous work of a real pricing page and onboarding flow.
The technical spine is usually Next.js on the frontend, Claude or OpenAI for reasoning and structured outputs, Supabase or MongoDB for state, Stripe for billing, and a lightweight prompt store so the product can improve without a full deploy.
Why founders lose months here
The AI SaaS market grew past 50 billion dollars in 2024 and analyst firms expect it to compound above 35 percent annually through the end of the decade. That has created a wave of technical founders trying to ship in categories they have never operated in before.
The most common failure I see is not the model. It is the six weeks lost to rebuilding auth, the four weeks lost to a self designed billing system, and the two months lost to an internal admin panel that never should have been built. I bring the discipline to skip that work and get to the actual product.
Build vs buy, seriously
The most valuable thing I bring to an AI product conversation is a strong opinion about what not to build. I have made those calls under my own name for my own products, and I bring the same discipline into client work.
Better Auth or Clerk for auth. Stripe for billing. Supabase for state. Cloudflare or Vercel for the edge. The differentiated work is the prompt design, the eval harness, and the interface. Everything else is a decision to keep off the roadmap.
How engagements work
Two common shapes. A fixed price MVP build that takes a founder from figma or a spec to a paying beta in four to eight weeks. Or a retained fractional CTO engagement where I ship alongside the founder for a defined number of days per month, usually four to eight.
Both start with a paid two hour architecture review. That is where the build versus buy calls get made and where the roadmap gets shortened.
Recently shipped
push8, the intelligence and management layer for n8n. Live at push8.dev in early access. Handles workflow building from plain English, silent failure detection, encrypted backups, and token usage tracking across OpenAI, Anthropic, Gemini, Mistral, and OpenRouter.
Extinde, a browser extension builder platform that lets non developers ship real Chrome extensions from a natural language brief.
ExcelErrorFinder, a focused utility product that finds and explains errors inside Excel workbooks, sold as a low friction SaaS product.
What does building an AI product involve beyond the LLM call?+
Auth, billing, quotas, prompt versioning, guardrails, error states, an honest pricing page, onboarding, and support tooling. The LLM call is often 5 percent of the work.
Which LLM should I build on?+
Claude for reasoning, long context work, and structured outputs. GPT 4 class models for wide integration and lower latency short prompts. The right answer depends on your product's failure cost more than benchmark scores.
How do you avoid vendor lock in on the model layer?+
Keep prompts, tools, and outputs in a thin abstraction so you can swap providers without rewriting product logic. Store prompt versions so you can A/B and roll back independently of the model.