Driven by curiosity and a passion for understanding complex systems, I draw on an interdisciplinary background in physics, computer engineering, machine learning, and full-stack software–hardware system design, together with 20+ years of impactful R&D experience across academia, open source, startups, and industry, to build efficient, self-optimizing systems and help organizations transform breakthrough ideas into impactful technologies, products, communities, and businesses.

Throughout my career, I have pioneered and advanced novel research and engineering directions years—and sometimes decades—before they became mainstream, transforming them into technologies, methodologies, organizations, and communities with lasting scientific and real-world impact.

Today, I apply this experience to self-optimizing AI infrastructure, full-stack AI/SW/HW co-design, autonomous R&D systems, digital twins, and efficient computing platforms, helping AI labs, startups, data-center teams, enterprises, and investors de-risk emerging technologies and accelerate scientific discovery, innovation, and market impact.

• Founder, Chief Scientist and Chief Architect of cTuning Labs and cTuning.org, and co-founder of MLCommons through the cTuning Foundation
• Former Strategic Advisor to Lumai (AI Infrastructure & Optical AI Acceleration); Head of AI Systems R&D at FlexAI; VP of MLOps at OctoAI (now part of NVIDIA); Founder and Chief Architect of the Collective Knowledge platform acquired by OctoAI; Co-Director of the Intel Exascale Lab; Senior Research Scientist at Inria; Adjunct Professor at Paris-Saclay University; technology architect and advisor on full-stack AI/SW/HW co-design and automation in projects with Arm, General Motors, Google, Qualcomm, and Amazon;
• Creator of the Collective Knowledge concept and platform, Collective Mind automation language, and MLPerf automations for reproducible experimentation, optimization, benchmarking, and full-stack software–hardware co-design of AI systems;
• Top open-community MLPerf inference submitter for several years and creator of automations that helped the community generate 10,000+ reproducible AI benchmark submissions and identify Pareto-efficient systems in terms of performance, cost, energy, and scalability across diverse cloud-to-edge CPUs, GPUs, and software stacks;
• Creator of the Open MLPerf Dataset and FlexBoard at Hugging Face, an open dataset and interactive analysis tool for using AI, data analytics, and reproducible experimentation to improve AI system design, deployment, performance, cost, and energy efficiency;
• ACM reproducibility and artifact evaluation leader; author of the unified Artifact Appendix and reproducibility checklist adopted and extended by major ACM and IEEE systems conferences
• ACM/IEEE Test of Time Awards (2017, 2025), h-index of 29, and 100+ publications and invited talks on self-optimizing compilers and computing systems, full-stack SW/HW co-design, and reproducibility
• PhD in Self-Optimizing Systems and Compilers from the University of Edinburgh

I’m currently focused on advancing the next generation of self-optimizing AI infrastructure, full-stack software–hardware co-design, and efficient computing systems while helping startups, enterprises, and investors identify and evaluate emerging opportunities, build high-impact R&D capabilities, accelerate innovation, and avoid costly technical and organizational pitfalls.

I’m an interdisciplinary scientist, systems architect, entrepreneur, educator, and R&D leader working at the intersection of physics, computer engineering, machine learning, and AI infrastructure. I apply first-principles thinking and data-driven optimization to transform breakthrough ideas and complex systems into impactful and scalable technologies, products, communities, and businesses. My work bridges cutting-edge research, engineering execution, and business outcomes to create intelligent systems that are efficient, scalable, cost-effective, and sustainable.

For more than two decades, I have repeatedly pioneered and advanced important ideas at the intersection of computing, physics, engineering, machine learning, and AI infrastructure—often years and sometimes decades before they became mainstream industry practices. Throughout my career, I have developed some of the earliest self-optimizing compilers and systems, pioneered machine-learning-guided optimization, reproducible and collaborative R&D, artifact evaluation, workflow and agent-based automation, knowledge management, and ML-driven software–hardware co-design. Several of these research directions later evolved into broader industry and research trends and have received international recognition, including ACM/IEEE Test of Time Awards for long-term scientific impact.

Among the early pioneers of artifact evaluation and reproducible systems research, I helped establish practices, infrastructure, and reporting methodologies that improved the quality, transparency, reproducibility, and long-term impact of computer systems and AI research. I also explored early forms of agent-based and autonomous research automation in my Collective Knowledge and Collective Mind projects years before the recent emergence of AI agents and AI-assisted scientific discovery.

My background spans academia, startups, open-source ecosystems, and industry leadership. I have served in roles including CEO, CTO, Head of R&D Lab, VP of MLOps, Chief Architect, Research Scientist, Systems Engineer, and Adjunct Lecturer. I founded deep-tech initiatives such as cTuning and Collective Knowledge to automate experimentation, benchmarking, knowledge management, reproducible R&D, and software–hardware co-design, accelerate collaborative R&D, and optimize modern AI and computing workloads across diverse hardware and software platforms.

From designing analog neural hardware in the 1990s to developing some of the first self-optimizing compiler stacks, inventing automation, benchmarking, and software–hardware co-design technologies adopted by MLCommons, and helping translate research innovations into commercial products, startups, and acquisition pathways, I have consistently worked at the intersection of scientific exploration and real-world impact.

In recent years, I have also advised startups, enterprises, and investors on deep-tech opportunities, technology strategy, technical due diligence, AI infrastructure, software–hardware co-design, and emerging technology trends. Drawing on experience across research, engineering, startups, and industry, I help organizations build high-impact R&D labs, avoid costly technical and organizational pitfalls, accelerate prototyping and validation, optimize performance, energy efficiency, and cost, and shorten the path from innovation to market impact.

My current interests include self-optimizing AI infrastructure, software–hardware co-design, autonomous R&D systems, digital twins, performance engineering, sustainable computing, and optimizing complex systems across the full stack to maximize performance, scalability, robustness, and scientific productivity while minimizing cost, energy consumption, complexity, and time-to-market.

I am passionate about building technologies, interdisciplinary teams, startups, and communities; mentoring researchers, engineers, and entrepreneurs; and helping organizations identify and execute on transformative opportunities in AI and computing to accelerate scientific discovery and technological innovation.

Selected impact:

• ACM/IEEE Test of Time Awards (2017, 2025)
• Co-founder of MLCommons through the cTuning Foundation
• Creator of Collective Knowledge, Collective Mind and MLPerf automations
• Collective Knowledge platform acquired by OctoAI (later NVIDIA)
• Author of the Artifact Appendix adopted and extended by major ACM and IEEE conferences
• 100+ publications and invited talks on AI systems, reproducibility, and self-optimizing computing

Key strengths:

• Interdisciplinary deep-tech R&D leadership
• AI infrastructure and ML-based software–hardware co-design
• First-principles system design and optimization
• Efficient computing, performance engineering, and cost optimization
• Technical due diligence and technology strategy
• Building R&D labs and innovation ecosystems
• Startup creation, scaling, and commercialization
• Open-source ecosystems and collaborative R&D
• Uniting and mentoring research, engineering, and business teams