Matteo Corbetta, Ph.D.
| GitHub | Semantic Scholar |
About
Applied ML Scientist with 10+ years of experience in data science and probabilistic modeling, ML, and AI for sensor-driven systems across industrial and safety-critical domains.
- Expertise in time-series anomaly detection, state estimation & sensor fusion, and forecasting for uncertainty-aware decision making
- Experience building models from low-TRL (fundamental research) to deploying C++ into production-grade environments
- Hands-on ML technical lead and individual contributor in research labs, large enterprises, and early-stage startups
- Core contributor to open-source research software, PI of funded research proposals at NASA, and recipient of multiple awards
Tech stack:
Python: PyTorch, Scikit-Learn, JAX, TensorFlow, Ray, Pandas, LangChain | SQL | C++ | Docker | AWS, GCP, Kubernetes.
Selected Accomplishments
Business Outcomes
- Increased reach of automotive wheel alignment monitoring system to more than 250,000 additional potential customers
- Led a team of 3 building a multi-agent workflow that enabled the onboarding of a new customer for GenAI, Series A startup
- Designed, implemented and deployed an AI-based root cause analysis of cost spikes for Cloud FinOps customer
Awards and Recognitions
- Core algorithms contributor to NASA’s 2024 Software of the Year: ProgPy, 2024
- OneKBR Award for outstanding work in the NASA’s Diagnostics and Prognostics Task, 2023
- Outstanding Reviewer for the Prognostics and Health Management Society, 2018
- Best paper award at European Prognostics and Health Management Conference, Bilbao, Spain 2016 (link)
- Best paper (3rd) at European Safety and Reliability Conference, Amsterdam, The Netherlands, 2013
Scientific and Technical Contributions
Awarded Project Proposals (PI or Co-PI)
- “Physics-aware quantum neural network modeling of Earth science phenomena”, NASA Ames AIST 2024
- “Acoustic Data-Based 0-gravity Boiling Characterization”, NASA Ames CIF 2023
- “Physics-Informed Neural Networks for Next-Gen Electric Aircraft”, NASA Ames CIF 2022
Conferences and Technical Societies
- Panelist at the SuperComputing Conf. 2022: “Physics-Informed Machine Learning meets High Performance Computing”
- Member of the Editorial Board for the Prognostic and Health Management Society 2017 - 2024
- Presented technical work at more than a dozen conferences and workshops.
- Reviewer for a number of scientific journals and conferences for more than a decade
Invention Disclosures and Patents
- Coolant Pump and Valve Prognostic Strategy (Ford Motor Company), 2024 (invention disclosure submitted)
- Bayesian network for fault isolation of UAV electrical powertrain (KBR & NASA), 2022 (patent pending)
- Vibration-based monitoring of wind turbine direct-drive generators (Siemens Gamesa), 2017 (invention disclosure)
Sample Work
Presentation: PIML for Prognostics and Health Management
Selected Publications
ProgPy: Python Packages for Prognostics and Health Management of Engineering Systems (2023)
C. Teubert, K. Jarvis, Matteo Corbetta, Chetan S. Kulkarni, M. Daigle
Journal of Open Source Software
Hybrid physics-informed neural networks for lithium-ion battery modeling and prognosis (2021)
R. Nascimento, Matteo Corbetta, Chetan S. Kulkarni, F. Viana
Journal of Power Sources
Application of sparse identification of nonlinear dynamics for physics-informed learning (2020)
Matteo Corbetta
IEEE Aerospace Conference
Comparison of Surrogate Modeling Techniques for Life Cycle Models of Advanced Air Mobility (2023)
A. Pohya, G. Wende, Matteo Corbetta, Chetan S. Kulkarni
AIAA AVIATION 2023 Forum
Particle filtering‐based adaptive training of neural networks for real‐time structural damage diagnosis and prognosis (2019)
F. Cadini, C. Sbarufatti, Matteo Corbetta, Francesco Cancelliere, M. Giglio
Structural Control & Health Monitoring
Hybrid Modeling of Unmanned Aerial Vehicle Electric Powertrain for Fault Detection and Diagnostics (2023)
Matteo Corbetta, K. Jarvis, S. Schuet
AIAA AVIATION 2023 Forum
Systems Health Monitoring: Integrating FMEA into Bayesian Networks (2021)
Chetan S. Kulkarni, Matteo Corbetta, E. Robinson
IEEE Aerospace Conference
Uncertainty Quantification of Expected Time-of-Arrival in UAV Flight Trajectory (2021)
P. Banerjee, Matteo Corbetta
AIAA AVIATION 2021 FORUM
On the performance of a cointegration-based approach for novelty detection in realistic fatigue crack growth scenarios (2019)
M. Salvetti, C. Sbarufatti, E. Cross, Matteo Corbetta, K. Worden, M. Giglio
Mechanical systems and signal processing
Approach for Uncertainty Quantification And Management Of Unmanned Aerial Vehicle Health (2019)
Matteo Corbetta, Chetan S. Kulkarni
Annual Conference of the PHM Society
Enabling in-time prognostics with surrogate modeling through physics-enhanced Dynamic Mode Decomposition method (2022)
K. Jarvis, Matteo Corbetta, C. Teubert, S. Schuet
Annual Conference of the PHM Society
Accelerating uncertainty propagation in power laws for prognostics and health management (2020)
Matteo Corbetta
IEEE Aerospace Conference
Optimal tuning of particle filtering random noise for monotonic degradation processes (2016)
Matteo Corbetta, C. Sbarufatti, M. Giglio
PHM Society European Conference