I'm a software engineer at Fondazione Bruno Kessler in Trento, Italy, in the Planning, Scheduling and Optimization unit. I build planners and schedulers, mostly in Rust and Python, and I care about making them fast and easy to use in real projects.
- tamerlite: a heuristic-search planner for temporal and numeric problems. It has a Rust core with Python bindings (PyO3), and it placed 3rd in the Satisficing subtrack of the Numeric Tracks at the International Planning Competition 2026.
- cpse: a constraint programming scheduling engine built on Google OR-Tools (CP-SAT). It plugs into Unified Planning as a regular planner.
- tampest: task and motion planning, which I helped extend to scheduling and motion planning. This work became an ICAPS 2026 paper.
- Unified Planning: an open-source Python library for automated planning. I've merged 60 pull requests there so far: compilers and the grounder, plan validators, motion planning support, and PDDL/ANML readers and writers.
- AI4Work schedulers: scheduling services for manufacturing, healthcare, and logistics pilots in a Horizon Europe project, running OR-Tools behind REST APIs in Docker (manufacturing, healthcare, logistics).
- pytorrent: a BitTorrent client in Python that I wrote to learn the protocol. It supports DHT, peer exchange, and UDP trackers.
- PlanSys2 + Unified Planning: ROS2 modules that let PlanSys2 use planners from Unified Planning.
Python · Rust · Java · Google OR-Tools · Docker · GitHub Actions / GitLab CI · ROS2
LinkedIn · MSc in Artificial Intelligence Systems, University of Trento