Daniele Moro
Chief Science Officer at Ironsite
Sunnyvale, California, United States
Summary
👤
Senior🎓
Top SchoolDaniele Moro is a Chief Science Officer and applied ML systems engineer with a decade of experience building ultra-efficient ML models and hardware-software co-design for production at Google, DeepMind, YouTube, and startups. He led teams that delivered orders-of-magnitude efficiency gains for LLMs and vision models—work that spawned papers like PikeLPN and drove a 3x energy reduction and, in one YouTube deployment, over 300x inference cost savings versus prior state-of-the-art. A hands-on researcher and backend developer, Daniele has contributed to the widely used ONOS open network OS (adding P4/UPF support and traffic metrics) and helped launch the world’s first transformer ASIC inference engine at Etched. Based in Sunnyvale, he blends academic publications and patents with scaled product delivery, and has a track record of turning research prototypes into production hardware-accelerated systems. Notably, he often focuses on quantization and co-design trade-offs that make large models viable on constrained hardware.
10 years of coding experience
12 years of employment as a software developer
Bachelor of Science - BS Computer Science, Bachelor of Science - BS Computer Science at Boise State UniversityTimberline High School
English, Italian, Spanish
Github Skills (5)
onos10
javas10
network-programming10
java10
protocol-buffers8
Programming languages (12)
Github contributions (5)
opennetworkinglab/onos
Dec 2017 - Apr 2022
Open Network Operating System
Role in this project:
Back-end Developer
Back-end DeveloperContributions:143 commits in 4 years 4 months
Contributions summary:Daniele made several significant contributions to the ONOS project, primarily focused on the core network operating system. They added support for suggested paths within the PointToPointIntent compiler and introduced functionality to clone packets to the CPU using clone sessions. Further work included implementing application filtering and various improvements related to UPF (User Plane Function) and P4Runtime, including the addition of the UpfProgrammable interface. Their work also involved adding metrics to monitor the traffic.
ospfopenflowbazeloperating-systemnfv
ANTLab-polimi/flowblaze.p4
Jul 2020 - Oct 2020
P4 library for experimenting with FlowBlaze and EFSM based stateful data planes
Contributions:95 commits, 8 PRs, 6 pushes in 3 months
statefulexperimentingplanes
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.