Consulting Software Engineer · AI & Data Engineering
Filippo Galli
I design, build and measure systems that combine software engineering, artificial intelligence and data — from RAG pipelines to healthcare applications.
I'm a software engineer with a solid academic background (MSc in Computer Engineering) and nearly two years of consulting experience in healthcare. Every day I work on real enterprise systems: HL7 FHIR APIs, the InterSystems IRIS multidimensional database, test automation and RAG-based assistants.
Beyond work, I build personal and university projects that let me explore what I'm passionate about: AI music generation, retrieval systems, developer tooling and real-time bots. I consider myself a fast learner who likes to measure their own results.
Featured projects
View all projects →RAG for InterSystems IRIS documentation
RAG backend for InterSystems IRIS docs: hybrid retrieval (dense + BM25), query expansion, cross-encoder reranking and grounded answers with validated citations. Offline retrieval evaluation included.
Drift — Emotion-Driven Ambient Music
The user's voice is analysed on two channels (prosody + semantics) to extract an emotion (valence/arousal), which drives a generative MIDI model and a real-time granular ambient synthesis.
Perceptor
Multi-process pipeline to capture and normalise chat, speech and video from live streams into timestamped JSONL events. VAD, speaker identification and ASR on CPU/GPU, VLM video captioning, all orchestrated over Redis queues.
Journey
See the full journey →Consulting Software Engineer
Healthy Reply · Padua (hybrid)
HL7 FHIR APIs on InterSystems IRIS, TrakCare EHR customisation, automated test suite (Java/Selenium/JDBC), internal RAG assistant for ObjectScript.
MSc in Computer Engineering
University of Padua
Thesis on emotion-conditioned sound generation (Deep Learning). CLEF 2024 competition with scientific publication.
BSc in Computer Engineering
University of Padua
Thesis on deep learning techniques for virus identification.