Hi, I am Luca and I work as software engineer at Eyecan.ai 🚀.
I am something between a scientist and an engineer, meaning that I often have to switch between pure software development and scientific research. A list of projects that I have worked on, including both development projects and research:
Pipelime is a pretty handy python library to automate data pipelines. We designed it to be as flexible as possible and to fit to every possible use case. It also comes with a DSL to post-process config files, which I personally maintain. Check it out!
Eyecandies is a fully synthetic, multi-modal, multi-light dataset for unsupervised anomaly detection with automatically labelled ground-truth annotations. It was published with an ACCV 2022 paper. Send us your results!
NNViz is a neural network visualization tool that can be used to quickly inspect the insides of a neural network. Currently works on pytorch models and renders them as PDFs. NNviz is borderline toy-project, but I plan to maintain it over time.
Pipewine is a complete rewrite of the Pipelime library aimed at improving various aspects of the original design including type safety, reducing dependencies, decoupling and rethinking various library components.