AI consulting and development services

I build, fix and lead AI products: RAG, LLM agents and the SaaS around them. Each offer below has a fixed shape and a typical budget, and the price is in writing before work starts.

Offers

• Build an AI SaaS MVP that holds up in production

  I build AI products from an empty repo to production in 8–14 weeks: retrieval, model calls, multi-tenancy, metering and billing. My own multi-tenant AI support platform, Evoriqa, went from first commit to live in about two months.

  Length

  8–14 weeks

  Typical budget

  $15k–$40k

• RAG development for a product you already run

  I add retrieval, an AI agent or an LLM feature to a product you already run, usually in 2–5 weeks. Answers come from your own content and say where they came from, the way AI PDF Chat cites the PDF and page.

  Length

  2–5 weeks

  Typical budget

  $4k–$10k

• Fractional CTO for AI startups and SaaS teams

  A fractional CTO, or a senior AI engineer, on retainer: the calls a CTO would make, without hiring one full-time. On most projects I am the only senior engineer, and I run my own AI SaaS in production, Evoriqa.

  Length

  Retainer

  Typical budget

  $4k–$8k a month

• RAG and LLM cost audit: accuracy and cost per request

  In 5–10 days I measure what your AI feature costs per request and how often it is right, then show which changes move both. On a CPA firm's review tool that measurement put a review at $0.24–0.28 and found a setting with more findings at 63% of the cost.

  Length

  5–10 days

  Typical budget

  $1.5k–$3k

• Take an AI prototype to production

  I take over half-built AI codebases: 2 weeks to stabilise, then ongoing work to extend them. The usual gaps are the same ones: no fallback when the API times out, no way to measure correctness, and nothing anyone can deploy.

  Length

  2 weeks to stabilise, then ongoing

  Typical budget

  $3k–$6k

• Modernise your SaaS without a rewrite

  I modernise SaaS products that have become hard to change, usually in 4–12 weeks, without pausing feature work. I start by learning why the system behaves the way it does, then change the parts that slow you down, one release at a time.

  Length

  4–12 weeks

  Typical budget

  $5k–$20k

How an engagement runs