Skip to content
View ameermasood's full-sized avatar

Block or report ameermasood

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.

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
AmeerMasood/README.md

Amir Masoud Almasi

Data Scientist | M.Sc. in Data Science @ Politecnico di Torino | B.Sc. in Mathematics | Ex-Data Analyst @ MCI (1+ year)


Tech Stack

Languages & Development    Python  |  C  |  SQL  |  Git  |  Streamlit  |  Gradio
Machine & Deep Learning    PyTorch  |  Hugging Face  |  Scikit-Learn  |  OpenCV
LLMs & Generative AI    LangChain  |  LangGraph  |  LlamaIndex  |  Ollama
Data Analytics & Visualization    Power BI  |  Excel  |  Pandas  |  NumPy  |  Matplotlib  |  Seaborn
Data Management & Big Data    PySpark  |  Hadoop  |  SQL Server  |  MongoDB  |  ChromaDB

Selected Projects

Python    |   LangChain    |   LangGraph    |   LlamaIndex    |   ChromaDB    |   DeepEval

  • Objective: Build a fact-checking system that can verify historical claims using RAG.
  • Architecture: Cyclic LangGraph workflow with specialized agents (Router, Librarian, Historian, Judge) over a ChromaDB + LlamaIndex retrieval stack with Nomic embeddings.

PyTorch    |   YOLO    |   ResNet    |   OpenCV    |   NumPy    |   Pandas

  • Objective: Texture-less 3D object pose estimation targeting robotics and autonomous manipulation.
  • Architecture: YOLOv10m detection combined with a ResNet18 heatmap regression network and multi-stage RGB-D cross-fusion.
  • Result: 92.4% mean ADD accuracy on LINEMOD, outperforming the RGB-only baseline by +7.9%.

PyTorch    |   Torchaudio    |   Hugging Face (Transformers, PEFT)    |   Gradio    |   HTML/CSS

  • Objective: High-fidelity speech emotion recognition evaluated on the complex ESD and IEMOCAP speech corpuses.
  • Architecture: Voxtral-mini-3B audio-language model fine-tuned with LoRA and DoRA PEFT on ESD & IEMOCAP.
  • Result: 0.84 macro-F1 in-domain and 0.63 zero-shot cross-domain generalization.

PyTorch    |   Hugging Face    |   RoBERTa    |   Adapters

  • Objective: Detecting sarcasm, irony, and sentiment under severe dialectal and domain shifts.
  • Architecture: RoBERTa-Large ensemble with Mixture-of-Adapters (MoA) and a Tensor-of-Cues (ToC) prompt-tuning mechanism.
  • Result: +16% cross-variety sarcasm classification and +18% zero-shot dialect adaptation.

Pinned Loading

  1. HeatNet HeatNet Public

    6D Pose Estimation via Keypoint Heatmap Regression with RGB-D Residual Neural Networks

    Python 6 2

  2. FigLangUnderstanding FigLangUnderstanding Public

    Figurative Language Understanding (FLU) System for Sarcasm and Sentiment Detection Using Transformers

    Python 1 3

  3. SpeechAgePrediction SpeechAgePrediction Public

    Age Estimation from Speech Signals via Time–Frequency Feature Extraction and Machine Learning

    Jupyter Notebook

  4. MultimodalSpeechEmotionRecognition MultimodalSpeechEmotionRecognition Public

    Speech Emotion Recognition Using Voxtral-mini-3B with LoRA/DoRA Fine-Tuning and Cross-Domain Evaluation on ESD and IEMOCAP

    Python 1