Contact me

David Joao David Joao, AI and automation Product Engineer in Évreux, Normandy and Paris

Évreux · Normandy · Paris

I turn ideas and business processes into smart, automated, production-ready digital products.

Contact me Explore my projects

You're not looking for one more tool. You're looking for someone who understands the problem, designs the solution, connects your systems and delivers a product that runs.

You are:

Where are things stuck today?

Tick what you're dealing with. I'll tell you what I would build.

Build. Automate. AI.
Or all three together.

Ten steps. One point of contact.

The full chain, from idea to live product. Every step produces a concrete deliverable.

The method in detail
  1. 01

    Idea

    We start from the business problem: who uses it, what costs time, what needs to change.

    DeliverableProduct brief and MVP scope

  2. 02

    UX/UI

    User journeys, key screens and a clickable prototype, approved before a single line of code.

    DeliverableApproved prototype

  3. 03

    Frontend

    A fast, responsive and accessible interface, built on a design system.

    DeliverableWeb application

  4. 04

    Backend

    Business rules, calculations and server-side processing.

    DeliverableApplication logic

  5. 05

    Database

    A clean, relational and secure Supabase data model.

    DeliverablePostgres schema and RLS rules

  6. 06

    Auth

    Accounts, roles and permissions. Everyone sees exactly what they should see.

    DeliverableSecure access

  7. 07

    API

    Connection to your existing tools through REST APIs and webhooks.

    DeliverableDocumented integrations

  8. 08

    Automations

    Repetitive tasks move into n8n workflows that run on their own.

    DeliverableWorkflows in production

  9. 09

    AI

    Assistants, agents, generation or classification, where AI genuinely saves time.

    DeliverableIntegrated AI features

  10. 10

    Go-live

    Deployment, domain, environments, monitoring and iterations.

    DeliverableLive product

The proof: systems that run.

Four reference systems to try out with sample data. Each one has its case study: problem, architecture, features, role.

AI qualification agent

Problem
Incoming requests arrive in a jumble and wait days for a reply.
System
The form triggers n8n, Claude analyses and scores the request, drafts a reply and creates the CRM record.
Stack
n8n, Claude, Supabase, webhooks
Read the case study

Automation workflow

Problem
Invoices, reminders and notifications handled by hand, every week.
System
Trigger, filters, database, AI and notifications chained in a workflow that runs on its own.
Stack
n8n, REST API, Supabase, Claude
Read the case study

Client portal with roles

Problem
Clients and teams communicate by email, WhatsApp and spreadsheets. Nobody knows where a request stands.
System
A portal with authentication, roles, request tracking and automatic notifications.
Stack
Supabase Auth, RLS, n8n, Vercel
Read the case study

Connected dashboard

Problem
The figures sit in five different tools and nobody has the full picture.
System
A dashboard wired to Supabase and your APIs, continuously up to date.
Stack
Supabase, REST API, n8n, Vercel
Read the case study
Qualification agent Sample data

Incoming requests

  1. Received by webhook
  2. Analysed by Claude
  3. Priority score
  4. Reply drafted
  5. CRM record created

Pick a request, then run the agent.

Workflow · Invoice reminders Simulation

0 run(s)

    Portal · Request tracking Sample data

      Dashboard · Operations Sample data
      Requests handled per day

      All case studies

      A stack chosen to ship fast and clean.

      Grouped by layer, from the interface down to the server.

      • Interface UX/UI, product prototyping, Lovable
      • Intelligence Claude, ChatGPT, Codex
      • Automation n8n
      • Integrations REST API, webhooks
      • Data Supabase: database, auth, storage
      • Delivery GitHub
      • Infrastructure Vercel, Hostinger, OVH, VPS

      AI-assisted engineering. Claude Code, Codex and Lovable to design and code faster, Higgsfield for visual creation.

      Portrait of David Joao in a dark jacket, hands clasped, backlit.

      I think product, I write the code, I connect the systems.

      A Product Engineer specialising in AI and automation, I work across the whole cycle: scoping, UX/UI, frontend, backend, database, authentication, APIs, automated workflows, AI features and deployment.

      Based in Évreux, Normandy, an hour from Paris, I work with SMEs, startups and teams on site in Normandy and Île-de-France, and remotely anywhere in France.

      I use Claude Code, Codex and Lovable to speed up engineering, never to replace thinking. What interests me: turning a complex idea into a simple tool that saves time, gives structure to an activity or automates a high-value task.

      Precise
      I show what is built, and why.
      Clear
      Technology explained without needless jargon.
      Results-driven
      The business problem comes before the tool.
      Transparent
      I separate what I have mastered, what I am experimenting with and what I am learning.

      I also run two digital agencies: Abraccio Studio in Luanda and Daelio in France.

      My approach in detail

      Which process do you want to improve, or which product do you want to launch? Let's talk.

      Contact me