Automate the work.Scale the business.

Ahmed Ben Yahia, AI automation engineer and full-stack developer
Ahmed Ben Yahia
Sfax, Tunisia · UTC+1
Available

Introduction

I’m an AI automation engineer and full-stack developer. I help operations, support and sales teams stop doing by hand what software can do reliably.

That means RAG knowledge assistants that answer from your documents, AI voice and chat agents that resolve customer requests, n8n workflows that move data between your tools, and custom SaaS or dashboards when off-the-shelf software no longer fits.

Every engagement is scoped in writing at a fixed price, delivered in weekly working increments and handed over with full documentation — so your team owns the system from day one.

systems shipped
+
client teams
+
on schedule
%

What I build

  • 01RAG knowledge assistants
  • 02AI voice & chat agents
  • 03n8n workflow automation
  • 04SaaS & internal tools
  • 05Lead & data pipelines
Book a diagnostic call

Free · 30 minutes · No obligation

Sectors served

Operations, support and sales teams in e-commerce, SaaS, property management, recruiting, field services, education and finance — wherever skilled people lose their week to repeatable work that AI and automation can take over.

02 — Practice areas

Six capabilities. One accountable engineer.

03 — How I work

A defined method, so you always know what happens next.

Read the full approach
  1. Phase 01Day 1

    Diagnose

    Map the workflow, quantify the manual effort, and define what success is measured against.

  2. Phase 02Day 2–3

    Specify

    A fixed-price proposal: scope, architecture, timeline, acceptance criteria and revision rounds.

  3. Phase 031–3 weeks

    Build

    Delivered in working modules, with a demo and written update every week.

  4. Phase 04Final week + 30 days

    Hand over

    QA, deployment, documentation and ownership transfer — followed by 30 days of support.

04 — Selected work

Real problems, the systems I built, and what changed.

Three engagements told the way clients ask about them: what was wrong, what I built, and the measurable result.

Case 01SaaS productSales & recruiting

Read case study

A LinkedIn outbound workspace

LinkedIn outbound product landing page

Challenge

Sales teams and recruiters ran LinkedIn outreach from spreadsheets, notes and disconnected tools, with no pipeline visibility and follow-ups slipping through.

What I built

A SaaS workspace for prospect lists, visual multi-step outreach flows, AI-assisted personalised messaging, team roles and campaign analytics.

Result

×

outreach capacity per team

  • Next.js
  • PostgreSQL
  • LinkedIn API
  • OpenAI

Case 02AI agentPrint & e-commerce

Read case study

An AI support desk across chat and phone

Support agent metrics dashboard

Challenge

More than 500 enquiries a day across chat and phone, a small support team, growing wait times and missed follow-ups.

What I built

One AI agent behind both channels — ElevenLabs voice over Twilio and web chat — with Stripe payments and booking inside the conversation.

Result

+

queries resolved daily · 45% faster replies

  • n8n
  • ElevenLabs
  • Twilio
  • OpenAI
  • Stripe

Case 03AI automationRecruiting operations

Read case study

A governed company-memory layer

Company-memory meeting intelligence settings

Challenge

Useful context from meetings, Slack and candidate conversations was lost — but raw data could not simply be dumped into a knowledge base.

What I built

A review-gated pipeline: meeting and CRM signals are extracted, approved in Slack, sanitised and committed to a Git-backed Markdown vault.

Result

%

of knowledge changes reviewed before they’re saved

  • n8n
  • Go
  • Next.js
  • Gemini
  • Docker

33 projects in the full portfolio.

12 automations4 AI agents10 SaaS products6 custom software builds1 desktop app

View all projects

05 — Is this a fit?

An honest answer before you book.

A strong fit if you…

  • Have a repeatable process eating 5+ hours of someone’s week
  • Want one accountable engineer instead of an agency hand-off chain
  • Value a fixed price and written acceptance criteria
  • Already use tools like HubSpot, Slack, Airtable or a CRM and want them connected
  • Want a system your team owns and understands after launch

Probably not a fit if you…

  • Need a large in-house team for a multi-year programme
  • Want an AI demo for a pitch, not a system that runs in production
  • Are comparing on the lowest hourly rate rather than a fixed, accountable outcome
  • Can’t make anyone available for a short weekly review and feedback
  • Haven’t yet decided which process matters most — though the free call can help with exactly that

The free diagnostic call

30 minutes. You leave with a plan — whether we work together or not.

  • Your biggest time sink, identified and sized
  • What to automate first — and what to leave alone
  • A realistic timeline and price range
  • No sales pitch, no obligation
Book your diagnostic call

06 — Ways to engage

Clear commercial terms from the first conversation.

Diagnostic call

30 minutes · No cost

We review the workflow, locate the real bottleneck and decide whether automation is worth it. You leave with a clear recommendation either way.

Fixed-scope project

Fixed price · From $300

Written scope, timeline and acceptance criteria. Includes build, testing, documentation and 30 days of support. AI platforms and SaaS are scoped individually.

Ongoing partnership

Monthly retainer

Maintenance, monitoring and new capabilities after launch, for teams that want a dependable engineer on call.

07 — Questions

Everything clients ask before booking.

Have a question that isn’t here?

Ask it on the free call, or send a message — you’ll get a straight answer within one business day.

Ask a question

Working together

Do I need a technical team to work with you?

No. Most clients are operations, support or sales leaders. I handle scoping, architecture and delivery, and explain every decision in plain business terms.

What do you need from us to get started?

Thirty minutes with the person who runs the process today, a few real examples of its inputs and outputs, and access to the tools the system will connect to.

Which time zones do you work with?

I’m based in Tunisia (UTC+1), with full overlap with Europe and morning overlap with North America. Replies come within one business day, usually much faster.

Pricing & timelines

How much does a project cost?

Every project is fixed-price, agreed in writing before work starts. Automation projects typically start from $300; AI platforms and SaaS products are scoped individually. You receive the price in your proposal within two to three days of our call.

How quickly will we see results?

You see working software in the first week. Automation pipelines and AI integrations usually go live in 1–2 weeks; full SaaS platforms or complex RAG systems in 3–6 weeks.

What if the result doesn’t match what we agreed?

Acceptance criteria are written into the proposal. If a deliverable doesn’t meet them, I fix it at no additional cost.

AI, data & ownership

Can an AI assistant give our customers wrong answers?

Assistants answer only from your approved sources, cite where each answer came from, and escalate anything they can’t ground to a person. They are tested on your real questions before launch.

Who owns the code, workflows and accounts?

You do. Everything is built in your accounts or handed over to them, with documentation and ownership notes, so you are never dependent on me to keep it running.

How is confidentiality handled?

An NDA is signed before anything sensitive is shared. Client data and workflows are never reused — which is why client names aren’t published on this site.

Will it work with the tools we already use?

Almost certainly. I’ve integrated 50+ platforms — HubSpot, Stripe, Twilio, Slack, Airtable, Supabase and more — and build around your stack rather than replacing it.

After launch

What happens if something breaks?

Every system ships with error handling, retries and alerts. Thirty days of support are included, and an optional monthly retainer covers monitoring, maintenance and new features.

Request a consultation

Tell me where the time is going.

Describe the workflow. I reply within one business day with initial thoughts and a time for a 30-minute diagnostic call.

Hours
Mon–Fri, 9:00–17:00 UTC+1
Area of interest
  • AI & knowledge systems
  • Workflow automation
  • AI voice & chat agents
  • SaaS & custom software
  • Lead & data pipelines
  • Not sure yet
Confidential. NDA available on request.