I hold a PhD in Computer Science and bring over 10 years of experience in software engineering, specializing in performance optimization and GPU-accelerated computing. I have a proven track record of driving significant improvements in query engine optimization, numerical computing, and deep learning, contributing to leading open-source projects such as PyTorch, Apache Arrow, NVIDIA RapidsAI, and more.
I’m currently a Staff Software Engineer at VoltronData, working with the Theseus team to develop a cutting-edge GPU Query Engine designed to handle petabyte-scale data. Theseus is optimized for massive ETL, SQL, and dataframe workloads, leaving CPUs behind for full native acceleration. We’re integrating advanced techniques from previous computing times to create the most powerful data engine in the world.
- ⚡ Advanced query optimization and indexing strategies, consistently delivering large improvements in query engine throughput and responsiveness.
- 🎯 GPU acceleration, leveraging CUDA and libraries like PyTorch and CuDF to achieve 10x-50x speedups in data-intensive workloads.
- 🔧 Pioneer in PyTorch sparse tensor support, enabling efficient processing of large-scale, sparse datasets.
- 🧠 Algorithm design, creating innovative solutions for complex data processing challenges in big data environments.
- PyTorch - Tensors and dynamic neural networks in Python with strong GPU acceleration.
- Apache Arrow - A development platform for in-memory analytics.
- CuDF - GPU DataFrame library.
- BlazingSQL - A lightweight, GPU-accelerated SQL engine for Python, built on RAPIDS cuDF.
I have a deep passion for data structures and algorithms. What fascinates me most is how they evolve and adapt as they scale—like building a complex bridge with Lego blocks, where each optimization introduces new strengths and challenges. If you’re passionate about these topics too, feel free to connect with me on X (formerly Twitter)!