Gabe Goodhart
Chief Architect AI Open Innovation
Denver, Colorado, United States
Summary
🤩
Rockstar🎓
Top SchoolGabe Goodhart is a seasoned software architect and AI leader with nine years at IBM where he has driven the consolidation of a unified watsonx AI platform across business units as Chief Architect, Open Innovation. He blends deep systems-level engineering—C/C++ inference work and SSL/server integrations for high-profile projects like llama.cpp—with cloud-native architecture and open source stewardship. Comfortable from one-off bash automations to designing abstract interfaces and logging frameworks, he consistently turns research-grade models into production-worthy services. Based in Denver, he pairs an MS in Computer Science with a background in computational science to bridge theory and pragmatic engineering. Uncommonly, he has contributed IBM Granite tensor support and model-conversion improvements upstream in LLM tooling, reflecting a knack for making cutting-edge architectures interoperable.
9 years of coding experience
12 years of employment as a software developer
BA, Computer Science, Mathematics, BA, Computer Science, Mathematics at Carleton CollegeMS, Computer Science, MS, Computer Science at University of North Carolina at Chapel Hill
Github Skills (11)
llama10
c-language10
ssl10
cmake10
cprogramming-language10
gfs9
python9
gf9
gff9
github-ci8
githubaction-workflow8
Programming languages (12)
Github contributions (5)
ggml-org/llama.cpp
Mar 2024 - Feb 2025
LLM inference in C/C++
Role in this project:
Back-end Developer
Back-end DeveloperContributions:11 reviews, 6 PRs, 30 comments in 11 months
Contributions summary:Gabe significantly contributed to the server-side functionality of the `llama.cpp` project, primarily focused on adding and configuring SSL support for the server, which included integrating CMake build toggles and setting up SSL key/cert file handling. They expanded the server's capabilities by incorporating IBM Granite architecture support, including new tensor definitions and modification of the model loading process. They also contributed to the template formatting for the IBM Granite architecture by including tool call formatting. Furthermore, the user adapted the conversion scripts for improved compatibility and proper configuration of XLMRoberta.
ggmlllama
caikit/caikit-tgis-backend
May 2023 - Aug 2024
Caikit module backend that manages models run in TGIS
Contributions:16 releases, 49 reviews, 33 PRs in 1 year 3 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.