
AI & Automation Engineer
IhebMarouani
I build agentic systems that do real work inside a business.
I came to AI throughcognitive science.
Studying how people actually reason turned out to be the most useful preparation possible for building systems that have to reason about a business. Today I am the first data and AI hire at an architectural glazing manufacturer, where I build the agentic systems its sales organisation runs on.
- Degree
- Cognitive Science B.Sc. (1.8)
- Thesis
- 1.0
- Languages
- German C1, English, French, Arabic
01In productionEmployer
ICP Engine & Data Enrichment
On-demand ideal-customer-profile engine built directly on live CRM data, enriched with verified online-identity and ad-behavior signals.
Python, Microsoft Dataverse, Embeddings, FastAPI
02In productionEmployer
Lead Generation Crawler
Agentic crawler that turns a filtered ICP into scored, source-attributed leads written back into the CRM.
Python, Agentic workflows, Dataverse
03PrototypeEmployer
Agentic Fleet
A fleet of domain-specialised AI agents over Microsoft 365. The company's first internal AI architecture, built with an evaluation method designed to prove whether it works.
Microsoft Copilot Studio, Power Automate, Dataverse, SharePoint, Entra ID
The whole floor.
Every screen down this avenue opens a prototype you can operate. Every figure, name and record inside those prototypes is invented; the projects behind them are not.
What I work with
- Languages
- Python, R
- AI & ML
- PyTorch, JAX, Flax, Claude API, LangGraph, Langfuse, Embeddings
- Data & backend
- Django, FastAPI, Flask, PostgreSQL, Neo4j, SQLAlchemy, Microsoft Dataverse
- Automation
- n8n, Power Automate, Microsoft Copilot Studio, crawl4ai, Apify
- Interface & deploy
- Astro, React, Streamlit, Cloudflare Pages, Railway
Open to rolesin Germany.
AI and automation engineering, and conversations about agentic systems in general.