Vinesh Janarthanan
Software Engineer - Build And Deploy at Cobot
California, United States
Summary
🤩
Rockstar🎓
Top SchoolVinesh Janarthanan is a software engineer specializing in build, deploy, and platform tooling with eight years of hands-on experience across large enterprise and open-source projects. Based in California and completing degrees in Computer Engineering and Computer Science at UW–Madison, he combines deep Bazel and CI/CD expertise with cloud deployments across Azure and AWS. At bp he shrank container images by 45%, automated complex Bazel pipelines, and drove test coverage and remote caching strategies for monorepo environments. He contributes to high-profile open-source ML infrastructure—enhancing llama.cpp with improved context management, a new Top-nσ sampler, and web UI fixes—bringing practical LLM inference improvements to C/C++ codebases. Comfortable across backend services, Python packaging, and frontend testing, he has a knack for turning performance and security requirements into repeatable pipelines. Colleagues rely on him to translate messy production problems into robust, testable build-and-deploy solutions.
8 years of coding experience
4 years of employment as a software developer
Computer Engineering, Computer Science, Computer Engineering, Computer Science at University of Wisconsin-Madison
Github Skills (11)
algorithm10
upsampling10
c-language10
gm10
gml10
subsampling10
sampling10
cprogramming-language10
sampler10
python5
json5
Programming languages (7)
Github contributions (5)
ggml-org/llama.cpp
Sep 2024 - Feb 2025
LLM inference in C/C++
Role in this project:
Back-end Developer
Back-end DeveloperContributions:11 reviews, 8 PRs, 41 comments in 5 months
Contributions summary:Vinesh contributed to the core functionality of the `llama.cpp` project, focusing on features related to context management and sampling strategies. They implemented an option to disable context shift, modifying the code to prevent context swapping under specific conditions. Furthermore, the user introduced a new Top-nσ sampler for token generation, enhancing the sampling capabilities and added tests. They also made changes to match OAI structured output response and introduced improvements to the web UI.
ggmlllama
VJHack/llama.cpp
Sep 2024 - Feb 2025
LLM inference in C/C++
Contributions:88 pushes, 12 branches in 5 months
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.