Stefano Fiorucci is a Senior Software NLP Engineer with 7 years of experience building production-ready language model applications and information extraction systems, currently contributing to deepset’s Haystack open-source framework. He blends back-end and DevOps skills—Dockerized Tika servers, crawler improvements, and an InMemoryKnowledgeGraph—with hands-on ML work using Hugging Face, Sentence Transformers, and vLLM. Previously at 01S he turned unstructured documents into accessible information for Italian citizens, and he maintains community-focused contributions to educational repos like mrdbourke/pytorch-deep-learning. Based in the Greater Perugia area, he pairs a civil engineering background and formal AI training with a practical curiosity for deployment, retrieval systems, and scalable LLM tooling.
7 years of coding experience
10 years of employment as a software developer
Professional Master Program Data Science, Professional Master Program Data Science at Università degli Studi di Perugia
School of Artificial Intelligence, School of Artificial Intelligence at Pi School
M2L - Mediterranean Machine Learning Summer School
AI orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your data. With advanced retrieval methods, it's best suited for building RAG, question answering, semantic search or conversational agent chatbots.
Role in this project:
Back-end & DevOps Engineer
Contributions:6 releases, 824 reviews, 41 commits in 8 months
Contributions summary:Stefano's contributions primarily focused on improving the infrastructure and adding features related to managing a Tika server, specifically within the context of an AI orchestration framework. They developed a `launch_tika` method using Docker to start a Tika server, and enhanced the crawler to extract hidden text from websites. Furthermore, the user added an `InMemoryKnowledgeGraph` to the codebase. The commits also involved improving code style and testing.
Materials for the Learn PyTorch for Deep Learning: Zero to Mastery course.
Role in this project:
ML Engineer
Contributions:2 commits, 4 PRs in 19 days
Contributions summary:Stefano primarily focused on improving and maintaining the exercise solutions within the `02_pytorch_classification_exercise_solutions.ipynb` file. This involved fixing typos, updating dependencies such as `torchmetrics` and adjusting the code to align with newer versions. Additionally, the user corrected the number of classes used in accuracy calculations.
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.