pietrobolcato / midas Goto Github PK

🎓 I am a Deep Learning Engineer, specialized in Generative AI (mostly NLP and CV) and MLOps

📜 BSc Software Engineering & MSc Artificial Intelligence, double degree from KTH Stockholm 🇸🇪 and TU Berlin 🇩🇪

❤️ I am passionate about Machine Learning/Deep Learning, SWE, Deploying models at scale, Generative AI, Computer Vision, NLP, Open source

🧠 Previously, I designed and deployed a text-to-sing system comprising four different deep learning models for realtime AI-generated melody, lyrics, and singing voice, used in the world’s most advanced AI Lyric Opera ever seen on a stage

🔥 Currenty working on a scalable voice-to-3D pipeline comprising of state-of-the-art deep learning models (Whisper, Stable Diffusion, ISNet and MiDaS) to generate in few seconds 3D content that can be placed in Augmented Reality by thousands of creators around the world

💡 Stack: Python, Pytorch, Terraform / Pulumi, Docker, Pytest, GH Actions, Azure (2x certified: Data scientist, Azure AI), AWS

🏠 Berlin, Germany 🇩🇪