Eric Lunderberg
MLSys Engineer at OctoML
Evanston, Illinois, United States
Summary
🤩
RockstarEric Lunderberg is an MLSys engineer with 13 years of experience, currently building production ML deployment and compilation tooling at OctoML from Evanston, Illinois. He blends low-level compiler and runtime expertise—demonstrated through contributions to Apache TVM—with practical ML systems work on projects like mlc-llm, where he optimized model prefill paths and quantization flows. Equally comfortable in systems and user-facing code, he has improved Rust terminal UI ergonomics in the popular ratatui crate by extending widget APIs for more concise usage. Eric is the kind of engineer who routinely moves between C++, Python, and Rust to close performance and usability gaps, and he often surfaces small but impactful refinements (e.g., arithmetic fixes and debug flags) that stabilize complex toolchains.
13 years of coding experience
Github Skills (29)
tui10
c-language10
language-model10
compilation10
python10
widgets10
datatable10
llm10
datatables10
compiler-design10
html-table10
compile10
terminal-ui10
rust10
cprogramming-language10
Programming languages (10)
Github contributions (5)
apache/tvm
Apr 2021 - Jan 2023
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
Backend Developer
Backend DeveloperContributions:925 reviews, 232 commits, 797 PRs in 1 year 9 months
Contributions summary:Eric has made multiple contributions to the Apache TVM project, focusing on the low-level details of compiler stack components. Their commits are centered on enhancing VTA runtime and arithmetic simplification rules, notably fixing zero-initialization and negative numerators. The user also contributed to the project's build process by removing unnecessary warnings for OpenCL wrappers. Their work involved file modifications in C++ and Python files, indicating a strong grasp of low-level compiler infrastructure and related tools.
metalvulkancompilertensoropencl
mlc-ai/mlc-llm
Aug 2023 - Jul 2024
Universal LLM Deployment Engine with ML Compilation
Role in this project:ML Engineer
ML EngineerContributions:15 reviews, 32 PRs, 38 comments in 11 months
Contributions summary:Eric made several contributions focused on optimizing the MLC-LLM deployment engine. They refactored code related to the Llama model, renaming variables to improve clarity and modifying functions in `mlc_llm/relax_model/llama.py`. The user also implemented a transform pass for applying split rotary embedding optimization to the prefill function. Additional contributions include adding a `--pdb` flag to build.py for debugging and cleaning up quantization-related code.
language-modelllmmachine-learning-compilationtvm
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.