Tobias Grosser
Associate Professor
Cambridge, England, United Kingdom
Summary
🤩
RockstarTobias Grosser is an associate professor and compiler researcher with 18 years of experience bridging academic leadership and low-level systems engineering across top institutions including Cambridge, Edinburgh, and ETH Zürich. He combines deep expertise in compilers, program analysis and formal methods with practical engineering—evidenced by performance-focused contributions to clang's Python bindings and test-suite enhancements in LLVM projects. Tobias has modernized ML-oriented code generation projects by removing legacy PPCG dependencies in TensorComprehensions and contributed formal bit-vector reasoning to the Lean 4 theorem prover. His work often targets behind-the-scenes bottlenecks—caching, direct lookups and translation-unit reparsing—to deliver measurable performance gains for developer tooling like clang_complete. Comfortable operating at the intersection of research and production, he brings a track record of refactoring complex codebases to be leaner, faster and more maintainable. Cambridge-based and research-active, he pairs rigorous formal thinking with hands-on optimization skills that improve both correctness and speed.
18 years of coding experience
6 years of employment as a software developer
Github Skills (33)
algorithm10
algorithms10
c-language10
performance-monitor10
python10
testing10
vimscript10
code-completion10
c1110
libclang10
refactor10
c1710
compiler-compiler10
performance-analysis10
optmization10
Programming languages (25)
Github contributions (5)
facebookresearch/TensorComprehensions
Feb 2018 - Aug 2018
A domain specific language to express machine learning workloads.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:532 commits in 6 months
Contributions summary:Tobias's contributions primarily revolved around refactoring and optimizing the code within the tensorcomprehensions repository, specifically targeting the removal of legacy dependencies related to PPCG (Polyhedral Parallel Code Generator) and associated wrappers, indicating a shift away from the older code generation method. They removed legacy code and deprecated functionalities (like MakePoint and EvalIntegerAt) related to PPCG integration to make the code more streamlined. The commits included fixes to comments and the modification of the codebase to use more streamlined methods and remove dependencies that were no longer in use.
workloadsexpressmachine-learningdomaindomain-specific
xavierd/clang_complete
Dec 2010 - Nov 2014
Vim plugin that use clang for completing C/C++ code.
Role in this project:Back-end Developer
Back-end DeveloperContributions:172 commits, 3 comments in 3 years 11 months
Contributions summary:Tobias primarily contributed to the development of a Vim plugin that utilizes the clang compiler for C/C++ code completion. Their work involved adding features to leverage libclang directly, improving performance, and integrating Python bindings for the clang library. They created functions to call the clang binary and libclang, incorporating diagnostics and code completion capabilities. The user also focused on optimizing the process by implementing features to reparse and cache existing translation units.
cppvim-pluginvimclang
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.