Skip to content
View Adriano-7's full-sized avatar

Highlights

  • Pro

Organizations

@NIAEFEUP

Block or report Adriano-7

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Adriano-7/README.md

Hi, im Adriano!

I recently finished my M.Sc. in Artificial Intelligence at the University of Porto. My thesis benchmarked open-weight LLMs as negotiation agents, and before that I researched how class imbalance degrades synthetic tabular data generation.

Alongside my studies, I helped redesign UNI, the app UPorto students use every day, and led the IT department at ESN Porto.

Some of my ML Projects

Project What it is Stack
Politeness Classification Four-way politeness classifier on Intel's Polite Guard dataset, SVM + Word2Vec baselines, RoBERTa fine-tuning, LoRA, and LLM prompting. PyTorch, Transformers, PEFT
Deepfake Detection & Generation Trained CNNs and ViTs for detecting AI-generated faces and GANs and diffusion models to generate them. PyTorch, Diffusers
Chessboard Digital Twin Turns a photo of a chessboard into a FEN string, comparing a classical OpenCV pipeline against YOLOv8 + U-Net. OpenCV, Ultralytics, PyTorch

GitHub Streak

Pinned Loading

  1. NIAEFEUP/uni NIAEFEUP/uni Public

    Mobile app designed to help students of the University of Porto to manage their academic life.

    Dart 66 18

  2. imbalance-effects-on-synthetic-data imbalance-effects-on-synthetic-data Public

    Jupyter Notebook

  3. msc-thesis-llm-negotiation msc-thesis-llm-negotiation Public

    Benchmarking open-weight LLMs (Gemma, Mistral, Qwen) in NegotiationArena games and testing two inference-time techniques, Self-Refine and team negotiation.

    Jupyter Notebook

  4. fcup-natural-language-processing fcup-natural-language-processing Public archive

    NLP project classifying text politeness using the Intel Polite Guard dataset. Approaches span from traditional ML (SVM, Word2Vec) to advanced Transformer fine-tuning (RoBERTa), PEFT (LoRA), and LLM…

    Jupyter Notebook

  5. fcup-computer-vision fcup-computer-vision Public archive

    Forked from Minipoloalex/chessboard-computer-vision

    Computer vision project analyzing chessboard images to extract piece positions, count pieces, and reconstruct the full game state (FEN) into a digital twin. It explores both classical CV techniques…

    Jupyter Notebook

  6. feup-ldts-packman feup-ldts-packman Public archive

    Java recreation of Pac-Man developed with a focus on Software Design Patterns and TDD. Created for the Software Design and Testing Laboratory course @FEUP

    Java 1