Services
Software engineering, with AI where it pays off
Three ways I usually help. Most projects mix them: an app that needs an AI feature, or a team that wants its agents to understand the codebase.
AI integrations & MCP servers
Connect your product's data and tools to Claude, ChatGPT and other AI agents, with clear permissions and guardrails.
Outcomes
- Your team can query internal data from the AI tools they already use
- Agents can take real, scoped actions in your systems
- Repetitive operations work (triage, reports, data entry) runs as reviewed agent workflows
What you get
- An MCP server (TypeScript or Python) with authentication and narrowly scoped tools
- Agent workflows built on the Claude API, with logging and human approval where it matters
- Docs, tests and a handover session for your team
AI-assisted engineering setup
Make Claude Code, Cursor and other coding agents genuinely useful in your codebase, not just autocomplete.
Outcomes
- Agents follow your conventions instead of generic ones
- Reviews, tests and quality-gate fixes take less of your engineers’ time
- New teammates get productive faster with an agent that knows the repo
What you get
- Repo rules (CLAUDE.md / AGENTS.md) that encode how your team actually works
- Agent skills for branch review, test generation, quality-gate fixes and scaffolding
- A walkthrough for the team and a short adoption playbook
Web & mobile apps with AI built in
From MVP to production: Angular, Ionic and React frontends with a solid, tested API behind them.
Outcomes
- A product your users can rely on, on the web and in the app stores
- AI features (assistants, smart search, automation) where they add real value
- A codebase your next engineer will be happy to inherit
What you get
- Angular (signals, standalone) or React web apps and admin portals
- Ionic / Capacitor mobile apps for iOS and Android
- NestJS or .NET APIs on PostgreSQL, with unit tests and CI
- Modernising existing Angular apps: standalone, signals, new control flow, esbuild
How an engagement works
From first call to handover
- Intro call.A free 30-minute call to understand the problem, your users and your constraints.
- Scoped proposal.A written scope with milestones, priced as a fixed project or a weekly rate.
- Build in the open.Small pull requests you can review, and a demo every week.
- Handover.Docs, tests and a walkthrough, so your team owns the result.
Not sure which one fits?
Describe the problem in a few lines and I’ll suggest the smallest useful first step.