Tiark Rompf

Tiark Rompf is an Assistant Professor at Purdue University, with research spanning programming languages, compilers, and systems. His work bridges domains including architecture, databases, machine learning, and AI through projects like Reachability Types and Rhyme.

• Co-director of the Purdue Center for Programming Principles and Software Systems (PurPL)
• Scientific Advisor at SambaNova Systems
• Previously a member of the Scala team at EPFL

His research focuses on:

• Runtime code generation and advanced compiler technology
• Expressive data-centric query languages (Rhyme, Datalog)
• Reachability type systems for memory safety and effect handling
• Metaprogramming and logical relations for formal verification

Recent publications highlight contributions to Datalog compilation (Flan), nested data structures (Rhyme), and polymorphic reachability types. He leads projects exploring:

• Compiler optimizations for emerging architectures (GPU, TPU, FPGA)
• ML-driven compiler improvements
• Secure multi-party computation via metaprogramming

Scientific awards include:

• NSF CAREER Award (2016)
• Google Faculty Research Awards (2017, 2018)
• DOE Early Career Research Award (2017)
• ACM SIGPLAN PL Software Award (2019)
• GPCE Test of Time Award (2020)

Students and alumni from his group have joined institutions like Databricks, DeepMind, Galois, and Meta.

He teaches advanced compiler courses including:

• CS 590 - Advanced Topics in Compilers
• CS 352 - Compilers
• CS 502 - Graduate Compilers