February 2025 Meetup

Sourav talked about Pandas and FireDucks!

Accelerate your pandas application with FireDucks; A multi-threaded DataFrame library with JIT Compiler - Sourav Saha

In this talk, I will introduce FireDucks, that comes with a multithreaded kernel backend written in C++, a JIT compiler to auto-detect and optimize the existing performance issues involved in a user program, and a pure python frontend highly compatible with pandas allowing a native pandas application to be executed as it is without any manual changes. With the promise of highly compatible pandas APIs and the revved-up performance, FireDucks can serve the demands of this digital age and transform the arduous task of data wrangling into a more efficient and less taxing endeavor. It can be simply installed on any Linux system using pip.

Sourav has 11 years of professional experience at NEC Corporation, Japan in the diverse fields of High-Performance Computing, Distributed Programming, and Data Science. He holds a bachelor’s degree in Computer Science and is currently affiliated with the Digital Technology Development Laboratories at NEC as a Senior Research Engineer. Blending the mixture of different niche technologies related to compiler framework, high performance computing, and multi-threaded programming, his team is developing a high-performance compiler accelerated DataFrame library, named FireDucks. He is mostly passionate about program optimization, public speaking and has been speaking in several technical conferences, meetups etc.