My background is in electrical and electronics engineering. I work on applied AI, computer vision, backend pipelines, and scientific computing by building systems around messy inputs, checkable outputs, and clear limits.
Road Risk Review is a private product beta for turning recorded road footage into event timelines and investigation packs. Its public evidence includes a 0.194 s mean conflict-time error across seven contact scenarios in a nine-clip controlled set.
Active product
A private beta for reviewing recorded road footage with computer vision, evaluation gates, evidence exports, and verification reports.
The product turns multiple visual cues into timestamped risk events, explains what contributed to each flag, and exports reviewable evidence such as per-frame data, labelled video, timelines, and investigation reports.
The implementation stays private while selected case-study baselines and deterministic scenarios remain public. Those artifacts gate the claims shown here and expose failures instead of hiding them.
Second product
A live weather prototype built around inspectable data.
Skywatch combines a seven-day forecast with server-side location privacy and an optional public-webcam cross-check. The companion page makes provenance, handling, and the visual check inspectable.
Inspectable path
v0.1 prototypeLocation
Resolved server-side; request IP discarded after lookup
Forecast
Open-Meteo data with its source and issue time visible
Visual check
Public webcam frames compared with brightness and edge signals
Applied AI for real-world review and operations workflows
Computer vision on recorded road and environmental footage
Backend pipelines, evaluation gates, and evidence exports
Satellite analytics, retrieval, and scientific computing
Earlier systems where I built technical tools around satellite tracking, data visualization, and scientific computing. I keep them here because they show how my work moved from isolated prototypes toward applied systems with verification, documentation, and usable outputs.
01
TLE parsing and validation, SGP4 propagation through a proven library, coordinate transforms, diagnostics, a two-body fallback, and an experimental PyTorch differentiable wrapper inspired by ESA dSGP4.
Started during my Master's degree programme in Communications Engineering (CE) at Politecnico di Torino. Archived; not actively maintained.
Project page02
A React and Three.js interface for 1,000-plus active satellites, paired with a FastAPI and Skyfield service for cached TLE data and 36-hour rise, peak, and set estimates.
Repository03
A reproducible GLOBE API fetch-and-clean pipeline with provenance filters, notebooks, mapping, search, and downloadable observation data.
Companion page04
A multilayer transfer-matrix simulation comparing sensor stacks, cross-validated against an established package with the disagreement from the cited reference documented explicitly.
Companion pageRemote or relocation-friendly opportunities in applied AI, backend, computer vision, and product engineering, especially tools for review, analysis, or operations workflows.
Start a conversation