November Event Recap

Thank you to everyone who joined us for our November event!

Thank you to everyone who joined us for PyData Pittsburgh’s November event, Bodo DataFrames: A Fast and Scalable HPC-Based Drop-In Replacement for Pandas. Last week’s meetup featured Ehsan Totoni and Scott Routledge from Bodo.ai, who introduced how Bodo DataFrames brings true HPC performance to the everyday Pandas workflow. By combining MPI-based parallel execution with an inferential JIT compiler and automatic query optimization, Bodo lets data teams scale familiar Python code to massive datasets without rewriting anything or adopting complex distributed frameworks.

The speakers also highlighted Bodo’s expanding ecosystem, including the AI Toolkit for scalable LLM generation, embeddings, and distributed PyTorch training, and new Iceberg support for versioned, metadata-rich dataset management. Together, these tools position Bodo as a unified, Python-first platform for large-scale ETL, AI pipelines, and modern data lake operations.

Check out the full presentation slides here!