Skip to content

Service · Web development

Next.js apps with AI features that actually ship

I build production web apps in Next.js, React, and TypeScript, and layer real AI on top of them. Most “AI developers” can't actually code, and most engineers who can haven't shipped real LLM features. I do both, with years of production work behind it.

Adding AI to something you already run? AI integration services. Based in Connecticut, working US Eastern hours with teams anywhere in the United States. Local business? Website design in Connecticut.

Book a call20 min

What you get

Built to ship, not to demo

Production web apps

Next.js (App Router), React, and TypeScript, built to ship and maintain, not throwaway prototypes. Five-plus years of production work behind it.

Real AI features

OpenAI and Anthropic APIs, retrieval-augmented generation, agents, and MCP servers wired into real products, not demos that fall over in front of users.

Full-stack, end to end

Auth, databases, payments, and transactional email. Clerk, Supabase, MongoDB, Drizzle, Stripe, Resend: whatever the build actually needs.

Deployed and observable

Shipped on Vercel with analytics and performance monitoring in place, so the app is live, fast, and measurable from day one.

Process

How an engagement runs

  1. Step 01

    Intro call

    Twenty minutes on what you are building. If I am not the right fit, I say so on the call.

  2. Step 02

    Written scope

    What gets built, in what order, what it costs, when it lands. You approve it before I start.

  3. Step 03

    Live preview

    Work lands on a preview URL as it happens, at whatever pace the project actually moves.

  4. Step 04

    Launch and handover

    Deployed, monitored, documented. The code is yours, with no dependency on me to keep it running.

Pricing

Ways to work together

Projects run on a fixed scope and a fixed price, agreed in writing before any work starts, so there is no invoice you did not see coming. Smaller jobs run by the hour. Both are fine. If you are still working out how to choose someone, I wrote a guide on hiring a freelance web developer.

Scoping sprint

$2,500one week

A week that produces an architecture, a build plan, and a fixed estimate. For when the idea is clear but the shape of the build is not.

  • Architecture and data model
  • Milestone plan with real dates
  • Fixed price for the build itself
  • Credited in full against a build booked within 30 days

Full build

Most common
$15,000from

Idea to launched product. The most common shape of work I take on.

  • Next.js, React, and TypeScript
  • Auth, database, payments, email
  • Deployed with analytics and monitoring
  • Documented handover, and the code is yours

AI feature on an existing product

$6,000from

An LLM feature inside a product you already run, built to survive real users, real costs, and real edge cases.

  • RAG, agents, or tool-calling
  • Cost controls and rate limiting
  • Evaluated before it ships
  • Wired into the stack you already have

By the hour

$90per hour

Template setup, customisation, fixes, and smaller jobs that do not need a scoped project. Bring what you already have.

  • Template and theme setup or customisation
  • Fixes, updates, and small features
  • Work on WordPress, Squarespace, or whatever you run
  • Billed against work you approve first

Ongoing development

$4,000per month

Continued work on a product that is already live: new features, performance, and the maintenance that keeps it shippable.

  • Part-time from $4,000 per month
  • Full-time at $12,000 per month
  • Features, performance, and maintenance

Not sure which one fits? That is what the intro call is for.

Stack

The tools I build with

  • Next.js
  • React
  • TypeScript
  • Vercel
  • Supabase
  • Firebase
  • Clerk
  • OpenAI
  • Cursor
  • Claude Code
  • Linear
  • Notion
  • GitHub
  • MongoDB
  • GSAP

FAQ

Common questions

What kind of projects do you take on?

Production web applications: SaaS products, internal tools, client portals, dashboards, and marketing sites with real functionality behind them. Next.js, React, and TypeScript, with AI features when the product genuinely needs them.

Do you work with existing codebases?

Yes. A lot of the work is adding to something that already exists, whether that is a new feature, an AI layer on a working product, or picking up a build someone else left unfinished.

What does “real AI features” actually mean?

Retrieval-augmented generation over your own data, agents that call tools, MCP servers, and streaming LLM responses wired into a production app with error handling, cost controls, and evaluation. Not a chat widget bolted onto a landing page.

How long does a build take?

It depends on scope, which is exactly why scope gets written down before anything starts. A focused AI feature on an existing app is usually weeks. A product from scratch is usually months. You get a timeline in the written scope rather than a guess on the call.

Do you work with clients outside Connecticut?

Yes. Most work is remote and spans US time zones. Connecticut is where I am based, not a limit on who I work with.

What happens on the intro call?

Twenty minutes. You describe what you are building and what is blocking you. I tell you how I would approach it, roughly what it takes, and whether I am the right person for it.

Have a build in mind?

A short call to talk through what you're building and whether I'm the right fit. If it isn't, I'll say so and point you somewhere better.