About
I'm Luigi Cruz, a Computer Engineer. I build GPU-accelerated systems that process signals from space. Currently based in the Bay Area, I work as a Staff Engineer at the SETI Institute, where I develop real-time signal processing pipelines for the Allen Telescope Array, a 42-antenna radio interferometer searching for technosignatures and studying cosmic phenomena like fast radio bursts. My work sits at the intersection of high-performance computing, software-defined radio, and astronomy, using CUDA and modern C++ to push terabits of data through GPUs in real-time.
What I do for work is essentially what I do for fun. I've been building software for accelerated computing, data visualization, satellite reception, and radio systems since my undergraduate days in Brazil, and those personal experiments eventually turned into open-source projects used by hobbyists and researchers worldwide.
To learn more about my work, visit the Works page, or feel free to contact me!
Timeline
2025 — Presented at NVIDIA GTC on using AI to search for extraterrestrial signals. Leading testing of the NVIDIA IGX Thor for the next-generation signal processing at the Allen Telescope Array. Launched Stelline, a modular toolbox for real-time astronomy pipelines on GPUs. Published peer-reviewed work on real-time deep learning for transient detection.
2024 — Became Staff Engineer at the SETI Institute. Developed the world's first real-time AI search pipeline using NVIDIA Holoscan, processing 86 gigabits per second of radio data per GPU.
2023 — Continued collaborating with the Allen Telescope Array as a consultant for UC Berkeley's Breakthrough Listen program. Presented at GNU Radio Conference on CyberEther's Metal backend and at C++Now on BLADE's architecture.
2022 — Officially joined the SETI Institute as an intern at Hat Creek, California. Completed Bachelor's degree in Computer Engineering. Began developing BLADE (Breakthrough Listen Accelerated DSP Engine), the GPU-based beamformer powering the Allen Telescope Array and many other telescopes around the world.
2021 — Started developing CyberEther, a multi-platform GPU-accelerated interface for compute-intensive pipelines supporting CUDA, Metal, Vulkan, and WebGPU.
2020 — Started volunteering at the SETI Institute as an Accelerated DSP Engineer, tasked with developing the Allen Telescope Array's first CUDA-based beamformer. Created Kimera, a low-latency hardware-accelerated video streaming tool during the pandemic. Published research on deep learning frame interpolation for weather satellite imagery.
2019 — Created PiSDR, a Raspbian-based Linux distribution pre-loaded with SDR and HAM software for Raspberry Pi. The project has since reached over 1,100 GitHub stars and 53,000+ downloads.
2018 — Presented at DEFCON's Cyberspectrum on reverse engineering weather satellites and at Campus Party Brazil on GOES HRIT reception. Published first academic papers on deep learning applied to agriculture and satellite imagery.
2017 — Co-founded the Weather Satellites Academic Team and began contributing to the Open Satellite Project, writing software to receive, demodulate, and decode signals from GOES geostationary weather satellites.
2016 — Joined the university's High Power Rocketry Team as Head of Avionics, designing avionics and ground systems for competition rockets.
2015 — Started my bachelor's degree in Computer Engineering at the Federal University of Technology - Paraná (UTFPR) in Brazil.