Crutcher Dunnavant
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolCrutcher Dunnavant is a senior software engineer and infrastructure leader with over 15 years tackling pernicious distributed systems and petabyte-scale storage and analytics. He has designed globally consistent systems at Google (CitC), built high-throughput petabyte pipelines at 3Scan, and recently applied vision-transformer boundary fusion and data-stack rebuilds for weather modeling at Atmo. Comfortable across languages and polyglot architectures, he pairs deep systems engineering with ML experience—contributing to projects like Facebook Research’s FairScale and Java LangChain tests. He prefers small, ambitious teams or early-stage roles (CTO/Eng Director), but also serves as a tech lead or principal IC for major projects that demand long-term scaling. Crutcher is available for short engagements to stabilize, refactor, and modernize struggling codebases, with a pragmatic focus on delivery and measurable impact. Based in San Francisco, he blends research-grade thinking with production-hardened execution.
11 years of coding experience
18 years of employment as a software developer
Masters of Science Computer Science, Masters of Science Computer Science at The University of Alabama
English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (13)
unit-testing10
javas10
data-parallel10
pytorch10
data-parallelism10
distributed-training10
java10
document-processing10
testing10
python9
documentation9
machine-learning9
cuda8
Programming languages (10)
Github contributions (5)
langchain4j/langchain4j
Dec 2023 - Feb 2024
Java version of LangChain
Role in this project:
Back-end Developer
Back-end DeveloperContributions:11 reviews, 66 PRs, 21 comments in 1 month
Contributions summary:Crutcher contributed to the `langchain4j/langchain4j` repository by implementing and testing new features. Specifically, the user added unit tests and documentation to the `Image` and `Document` classes. Additionally, the user implemented several test classes to validate the functionalities of the `DocumentSplitter`, `DocumentLoader`, `DocumentTransformer`, and other core components, improving the code coverage of the project.
anthropicchatgptchromaembeddingsgemini
facebookresearch/fairscale
May 2022 - Aug 2022
PyTorch extensions for high performance and large scale training.
Role in this project:ML Engineer
ML EngineerContributions:11 reviews, 10 commits, 22 PRs in 2 months
Contributions summary:Crutcher primarily contributes to the testing and maintenance of the FairScale library, focused on PyTorch extensions for large-scale training. They fixed typos in documentation and removed false tuples within test files. Their contributions include modifications to the test suite by adding and updating test cases related to SSD offload, sharded data parallel, and checkpointing functionalities. The user also refactored and moved code for improved modularity and organization.
pytorchdeep-learningmachine-learningscaletraining
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.