Ongoing research
ParancU 2.0
Research into factual representations, question-oriented retrieval and answer containment: whether a retrieved passage contains the information needed to answer a question.
Maurizio Scibilia · AI Engineer
AI Engineer focused on retrieval, evidence and applied AI.
I'm Maurizio Scibilia, a software engineer and AI researcher with experience in software development, data science and machine learning. I founded ParancU Lab, an independent research initiative investigating how AI can identify factual evidence in documents, beyond semantic similarity alone. My work combines practical implementation, experimental evaluation and analysis of system limitations.
Released application · Available on the App Store
ParancU for iOS
I developed and released ParancU for iOS, a mobile application featuring AI-assisted document preparation and on-device multilingual retrieval.
Building on the EcoSearch research prototype, I took the project through research framing, architecture, implementation, benchmark design, failure analysis, mobile integration and App Store release. Corpus persistence, multilingual E5 inference, scoring, ranking and evidence inspection run on-device, with lexical signals available when exact or rare terms matter.
The v1.1 white paper documents the implementation, evaluation results and remaining limitations.
Experimental web application
ParancU Agent Lab
I subsequently developed ParancU Agent Lab, an experimental web application exploring document retrieval, AI-assisted generation and verification workflows.
Prepared ParancU corpora can be searched directly in the browser without an API key. Optional server-assisted generation and verification steps provide a way to investigate how retrieved passages support an answer and where information may be missing. These workflows remain experimental and do not guarantee factual correctness or complete evidence.
Current research at ParancU Lab
Alongside application development, I investigate document retrieval, factual evidence and evidence sufficiency. These are ongoing research directions; detailed explanations and updates are on parancu.com.
Ongoing research
ParancU Decision Lab
Research into evidence sufficiency, missing-information diagnosis and complementary evidence selection: how to assess what is available and identify what is still needed.
From EcoSearch to ParancU
The project evolved by questioning the retrieval objective itself.
Earlier work
EcoSearch
EcoSearch began as a retrieval-first QA prototype designed to make supporting passages visible before generation.
It became a test bed for chunking, semantic search, lexical signals, reranking, confidence routing and retrieval evaluation.
Research context
Evidence-oriented retrieval
Those experiments exposed a more fundamental question: semantic proximity is useful, but does it reliably identify the passage containing the evidence needed to answer a question?
ParancU focuses directly on that retrieval problem, independently from downstream language generation.
Professional background
Experience across software engineering, consulting, web development, data science and AI evaluation, including work with Siemens, Framfab / A.T. Kearney Applications & Technology, Deep Consulting, Spindox and others.
Engineering
Systems and implementation
Hands-on experience designing and building technical systems across different generations of software, from web and application development to modern AI and retrieval pipelines.
Evaluation
Measured behaviour
Current work places particular emphasis on benchmark design, failure analysis, retrieval quality and the distinction between plausible output and inspectable evidence.
Education & credentials
MSc in Computer Science
University of Pisa · A (highest honours). 18-month research thesis on LRAM graph neural networks, including original theorems on asymptotic and exponential stability.
Advanced education
Master in ICT, Cefriel / Politecnico di Milano · A · Scholarship recipient.
MBA, LUISS Guido Carli · A · Scholarship recipient, with international exchange at Paris Dauphine University.
TensorFlow Developer Certificate
2021–2024
Google Cloud Professional Machine Learning Engineer
2023–2025
Areas of focus
AI engineering with a strong emphasis on retrieval quality, evidence visibility and measurable system behaviour.
Information Retrieval
Evidence Retrieval
AI Evaluation
Natural Language Processing
Local AI
Embeddings
Retrieval Evaluation
Question Answering
Multilingual Systems
Applied AI Systems
Agentic AI
LangGraph
Grounded Generation
Professional profile
Research-minded, implementation-driven.
My work combines research framing with hands-on engineering: defining the problem, building the system, constructing evaluations, analysing failure modes and iterating until the behaviour becomes measurable and useful.
For full professional background, education, certifications and contact details, see my CV.