AI & Machine Learning Engineer | Python Developer
Building practical AI applications, speech processing pipelines, computer vision systems, and open-source developer tools.
I'm Reza Souri, a Python-focused AI and software engineer interested in bringing machine learning research into usable applications. My work spans deep learning, computer vision, natural language processing (NLP), speech and audio AI, and modular software engineering.
I care about reproducible experiments, clear APIs, maintainable pipelines, and tools that other developers can run and extend. I'm also exploring Retrieval-Augmented Generation (RAG) and large language model (LLM) application architectures.
A local-first desktop application and Python CLI for audio ingestion, vocal separation, speaker diarization, and speaker-labeled audio dataset preparation. Built around Python, PySide6/Qt, and machine learning models for speech processing.
Topics: Speech AI · Speaker Diarization · Audio Separation · Python · PySide6
An aspect-based sentiment analysis and review extraction project with reproducible benchmark preparation, evidence-span validation, and a local inference pilot. It emphasizes evaluation and data quality rather than unverified model-accuracy claims.
Topics: Natural Language Processing · Aspect-Based Sentiment Analysis · LLM Evaluation · Python
A video-processing workflow that uses face detection and recognition components across FastAPI microservices, with Docker-based deployment tooling.
Topics: Computer Vision · Deep Learning · Face Detection · FastAPI · Docker
My fork of the open-source Graphify project, where I'm exploring 3D graph visualization for HTML-based graph views and developer workflows.
Topics: Open Source · Graph Visualization · Python · Developer Tools
| Area | Technologies & interests |
|---|---|
| Machine learning & deep learning | PyTorch, TensorFlow, scikit-learn, model evaluation |
| Computer vision & data | OpenCV, NumPy, pandas, image and video processing |
| NLP, LLMs & RAG | Hugging Face Transformers, NLP pipelines, retrieval-augmented generation |
| Speech & audio AI | Speaker diarization, audio separation, dataset preparation |
| Software engineering | Python, FastAPI, PySide6/Qt, Git, Docker, GitHub Actions, Linux |
I enjoy developing practical tools, experimenting with new AI workflows, and contributing to open-source software. Browse my GitHub repositories to see more Python, ML, and software projects.
Have a question about a project or want to collaborate on AI engineering, computer vision, or speech processing? Reach out on LinkedIn or by email.