π― Currently looking for a final-year internship (PFE) 2026 β 2027 in AI, computer vision, data or software engineering β a pre-hire opportunity is very welcome.
Third-year Applied Computer Science engineering student at ENISo (Sousse, Tunisia). I build systems where computer vision, generative AI and software engineering meet: real-time video analytics, RAG pipelines, industrial IoT supervision and smart-city platforms β from the embedded sensor to the dashboard.
What I work on
- ποΈ Computer Vision & Deep Learning β YOLOv8 Β· DeepSORT Β· LSTM Β· OpenCV Β· TensorFlow Β· PyTorch
- π§ Generative AI β RAG pipelines Β· ChromaDB Β· Ollama Β· Groq Β· n8n automation
- π» Full-Stack Engineering β React / Next.js Β· Node.js Β· FastAPI / Flask Β· PostgreSQL Β· Docker
- π Industrial IoT & Automation β Delta PLC Β· Modbus RTU/TCP Β· ESP32 / STM32 Β· real-time supervision
Quick facts
- π Engineering cycle, Applied Computer Science β ENISo, Sousse Β·
2024 β 2027 - π Based in Monastir, Tunisia Β· open to relocation
- π Arabic (native) Β· French (B2) Β· English (B2)
- π€ ENISo Team Β· IEEE ENISo Student Branch Β· Orange Tech Club
- βοΈ firasmrabet1603@gmail.com
π What is inside each project
Distributed architecture: a Node.js backend parses native MJPEG streams and pushes low-latency video to a React dashboard over WebSockets. An AI pipeline (YOLOv8 Β· DeepSORT Β· LSTM) runs out-of-process through IPC for weapon detection, pose-based threat recognition and behavioural analysis, with human-verified alerts, Docker deployment and multi-channel emergency notifications.
Node.js React Python YOLOv8 LSTM WebSockets Docker
Full IoT + Edge AI stack for an industrial conveyor driven by two stepper motors: ESP32 sensor acquisition over MQTT, anomaly detection, fault classification and remaining-useful-life prediction, prescriptive maintenance recommendations and an 8-page real-time control room. Runs out of the box in simulation mode, then switches to real hardware without a code change.
FastAPI Next.js MQTT TinyML Docker
LAN/WiFi diagnostics with hardware-level discovery (ARP, 802.11 BSSID), interactive 3D topology in Three.js with live data-flow animations, a multi-provider LLM routing engine (8 providers) and PostgreSQL telemetry analytics.
Next.js FastAPI Three.js PostgreSQL Multi-LLM
French natural-language β SQL compiler, finite-state automata (DFA) modelling the lifecycle of urban entities, and a generative-AI module for reporting and intervention validation. Academic project β Language Theory & Compilation, ENISo.
Python Streamlit React MySQL Ollama
End-to-end RAG pipeline for investment scoring: multi-source extraction, chunking, ChromaDB retrieval, Groq LLM scoring, Plotly gauges and JSON export β the same workflow automated in n8n with Google Drive ingestion and LLM fallback.
Python Streamlit ChromaDB Groq n8n
Supervision of a denim chemical-dosing line: live tank levels, pressures, temperatures and valve states, manual/automatic control, Python anomaly detection and predictive maintenance, history and reporting. Built from a Delta DVP PLC + DOPSoft HMI through a Node.js Modbus RTU gateway.
React Node.js Python Modbus SQLite
π¬ Languages
π¨ Frontend
βοΈ Backend & Mobile
π€ AI, ML & Simulation
ποΈ Databases
π οΈ Tools, DevOps & Embedded
Also Β· YOLOv8 Β· DeepSORT Β· LSTM Β· RAG Β· ChromaDB Β· Ollama Β· Groq API Β· n8n Β· Streamlit Β· Pandas Β· Webots Β· ESP32 Β· STM32 Β· Modbus RTU/TCP Β· Delta PLC (ISPSoft) Β· DOPSoft Β· TIA Portal Β· WebSockets
AI & Industrial IT Engineering Intern Β· Sartex Group β Ksar Hellal, Tunisia
Jul 2026 β Aug 2026
- Automated a chemical dosing system on a Delta DVP PLC (Ladder / ISPSoft) with a DOPSoft touch HMI.
- Built a real-time React supervision dashboard connected to the PLC through a Node.js gateway (Modbus RTU, WebSockets, SQLite).
- Shipped a Python/Flask microservice with Z-score anomaly detection and linear-regression predictive maintenance.
Full-Stack Developer Intern Β· Bedouielec Transformateurs β Teboulba, Tunisia
Jul 2025 β Aug 2025
- Delivered a B2B e-commerce platform with React, TypeScript, Node.js/Express and Supabase.
- Integrated an AI conversational assistant for product search, personalised recommendations and chat-based add-to-cart.
- Added catalogue management, PDF quote generation, transactional emails and Docker containerisation.
| Engineering cycle β Applied Computer Science | ENISo, National Engineering School of Sousse Β· 2024 β 2027 |
| Preparatory cycle β Physics & Chemistry | IPEIM, Monastir Β· 2022 β 2024 |
| Baccalaureate β Experimental Sciences | LycΓ©e Teboulba Β· 2018 β 2022 |
- Building LLM Applications With Prompt Engineering β NVIDIA Β·
2026 - Agents Course β Certificate of Excellence β Hugging Face Β·
2026 - Oracle Cloud Infrastructure 2025 Certified AI Foundations Associate β Oracle Β·
2026 - Introduction to Internet of Things β Cisco Networking Academy Β·
2026
I am currently looking for a final-year project internship (PFE 2026 β 2027) in AI, computer vision, data or software engineering β and a pre-hire opportunity is very welcome. Feel free to reach out in English or French.









