Marco Mongi

License Plate Detector

Real‑time pipeline for license‑plate detection, multi‑object tracking and OCR, with an interactive UI and a fully reproducible, production‑oriented setup.

Try online →

Household Energy Forecasting

Forecasts household electricity consumption using a complete data‑science pipeline—from exploration and preprocessing to model training, comparison, and rigorous evaluation—aimed at improving planning and operational efficiency.

A report in scientific format and presentation are available (Spanish).

• EDA & ETL
• Feature engineering
• Time‑series models: FFNN, Prophet, RF
• Fourier seasonality
• Gradio notebooks
• LaTeX reporting
• Jupyter / Colab

Interactive Adventure Generator

Agentic storytelling app where an LLM narrates a branching adventure and adapts the plot to the player’s choices. Supports bilingual (EN, SPA) voice & text interaction, streaming responses, and configurable model providers.

• LLM API interaction
• Prompting
• Agentic flow (LangChain)
• TTS (Piper)
• STT (Whisper)
• Gradio UI

Blood Cells Count Aid

Automated microscopy assistant that detects and counts red blood cells, white blood cells, and platelets in stained blood-smear images, overlaying colour-coded markers in a PyQt5 desktop UI—achieving ≈ 98 % overall counting accuracy versus expert annotations on selected test images.

Market-Sentiment Predictor

Pipeline that scrapes finance-related tweets, extracts daily keyword sentiment with RoBERTa (≈ 90 % accuracy), fuses it with index prices to build time-series features, and trains classification/regression models for next-day moves—revealing only 50 % up-vs-down accuracy and underscoring the random-walk nature of the target.

OEE Analytics

End-to-end vision solution for a commercial bakery: smart cameras with object detection algorithms and DeepSORT tracking on Raspberry Pi 5 + Hailo edge HW stream MQTT counts to a Grafana OEE dashboard—boosting data-driven productivity across the line.

SABIAMar L0 Processor

Upgraded the Level-0 processor—the software responsible for decoding the raw binary image data transmitted by the SABIAMar Earth-observation satellite. Specifically, I implemented the unpacking of data frames across multiple levels—from transport and network protocols, through image and line layers, down to individual pixels—and added telemetry metadata to the resulting files. The Level 0 processor is the first and essential stage in a chain of processors that ultimately produces usable satellite images.

• Python
• Automated testing
• Binary-frame decoding
• Data-pipeline design

Satellite Crop Detector

Python platform that classifies crop types from time-series of multi-spectral satellite imagery: per-pixel reflectance series are processed with classical computer-vision techniques, engineered into features, and fed to a Random Forest model—reaching a 93 % F1 score.

Poultry Environment Regulator

Automated control system for chick brooders, using a microcontroller to manage heating, ventilation, and lighting schedules. The system is based on an Arduino UNO with DHT11 sensing, RTC time-keeping, and relay-controlled loads. Settings are adjusted via an LCD-encoder menu, with safe defaults restored after power loss.

Tablet-Joystick Dock

Custom 3D-printed add-on that clamps a joystick to a tablet while rerouting power, audio and volume controls through internal USB / 3.5 mm extensions and a dual-button circuit—restoring full charging, headphone and volume functionality during gameplay.

• 3D design
• 3D printing
• Electronics prototyping