Michele Pangrazzi
Senior Software Engineer at deepset
Trentino-Alto Adige/Südtirol, Italy
Summary
🤩
Rockstar🎓
Top SchoolMichele Pangrazzi is a Senior Software Engineer with 12 years of experience building production-ready backend systems and ML-driven NLP platforms from Trentino-Alto Adige, Italy. He has led migrations from rule-based to transformer-centric stacks, improving aspect-based sentiment analysis F1 by up to 40% while reducing maintenance costs and scaling pipelines on AWS. As an early R&D hire and technical lead, he designed core GraphQL APIs, data/ML pipelines using Luigi and ECS, and mentored teams in TDD and pair programming. Michele contributes to notable open-source projects in the AI/ML ecosystem—helping improve deepset’s Haystack retrieval and Hugging Face’s SetFit few-shot tooling—and has hands-on experience modernizing database integrations like MongoDB. His background blends systems engineering (CQRS, real-time monitoring, authentication at scale) with applied ML, making him comfortable moving models into robust, observable production services. Notably, he balances deep technical implementation with pragmatic product impact, having shipped solutions that processed millions of feedback items and supported multi-million user flows.
12 years of coding experience
16 years of employment as a software developer
Specialized education on ITC Technologies Telecommunications Informatics, Specialized education on ITC Technologies Telecommunications Informatics at Università di Trento
Italian, English
Github Skills (17)
transformers10
pytorch10
javascript10
sentence-transformers10
python10
machine-learning10
large-language-models10
mongodb-database10
retrieval-augmented-generation10
few-shot-learning10
nlp10
haystack10
mongodb10
testing9
generative-ai9
Programming languages (7)
Github contributions (5)
Automattic/monk
May 2019 - May 2019
The wise MongoDB API
Role in this project:
Back-end Developer / Database Engineer
Back-end Developer / Database EngineerContributions:15 commits, 2 PRs, 6 comments in 11 days
Contributions summary:Michele primarily focused on updating the project to accommodate the latest MongoDB versions, as evidenced by the "update mongodb to 3.x" commit. This included changes to database interactions, specifically around index handling and deprecation notices related to the `geoNear` command. The user also addressed deprecation warnings and added new features related to querying options.
apimongodb-apiwisenodejsmongoose
deepset-ai/haystack
Apr 2022 - Apr 2022
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 Developer
Back-end DeveloperContributions:60 reviews, 1 commit, 18 PRs in 1 day
Contributions summary:Michele primarily contributed to the Haystack framework by addressing issues related to the `EmbeddingRetriever` and related components. They implemented a warning message to alert users when a Sentence Transformer model is used with an incorrect model format. Additionally, they removed deprecated parameters from components such as `PyPDFToDocument` and updated the default values of the `store_full_path` parameter within the converters. The user also worked on the serialization and deserialization of the NamedEntityExtractor component.
pythonbertfinetuningproduction-readyhuggingface-transformers
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.