Looking for a job

Neirouz Adili

@neirouzadili

AI engineering student at ESPRIT building forecasting models deployed through Orange RDC APIs and agentic AI systems for applied machine learning.

Tunisia

Message

What I'm looking for

I’m seeking a final-year internship in recommendation and applied AI, available from January 2027.

At Orange RDC, I developed a LightGBM traffic forecasting pipeline that predicted congestion six months ahead and was benchmarked against a Holt-Winters baseline. I also deployed a real-time inference API and built a Streamlit decision-support simulator.

At ESPRIT’s Digital Learning Platform, I designed a Bloom’s taxonomy-based multi-agent system to assess learners’ cognitive levels. On a separate cybersecurity internship at iMedrasphere, I developed an XGBoost phishing classifier exposed through a Flask REST API.

My projects include CV Ranker, an explainable CV-to-job matching tool, and MindFlow – EVO, an agentic study companion with a LiteLLM multi-LLM router and a Raspberry Pi 5 robot connected over MQTT. I’m completing an engineering degree in Internet of Things at ESPRIT and am seeking a final-year internship in recommendation and applied AI, available from January 2027.

Experience

Work history, roles, and key accomplishments

AI Intern

ESPRIT

Jul 2026 - Aug 2026 (1 month)

Designed a multi-agent system based on Bloom's taxonomy to assess learners' cognitive levels, increasing engagement by 15% and evaluation accuracy by 20%.

Multi Agent Systems LangGraph RAG LLM Edtech

Network Automation & AI Intern

Orange RDC

Jun 2026 - Aug 2026 (2 months)

Developed a traffic forecasting pipeline using LightGBM, achieving a MAE of 3.17 and 65% better accuracy than a Holt-Winters baseline. Deployed a real-time inference API with FastAPI and a decision-support simulator with Streamlit.

LightGBM fastAPI Streamlit Time Series Feature Engineering Benchmarking Monte Carlo Simulations REST APIs

Cybersecurity Intern

iMedrasphere

Jul 2025 - Aug 2025 (1 month)

Built an XGBoost phishing detection classifier with 94.2% precision and 0.98 ROC-AUC, exposed via a Flask REST API for real-time scoring, replacing a static blacklist.

XGBoost Flask REST APIs Feature Engineering Cross Validation Cybersecurity