Ricky Loynd is an experienced AI researcher and engineer with over 10 years building language models, transformer-based RL agents, and long-term memory agents at Microsoft Research and as an independent researcher. He architected large-scale language modeling systems and concept-extraction pipelines that were integrated into Bing and Microsoft Cognitive Services, and contributed core speech recognition innovations that helped make voice a standard feature in Windows. Ricky combines deep research instincts with hands-on software development—from N-gram lookup services and entity linking to practical ML systems—and has a track record of shipping production-grade components and tooling. He is an active contributor to prominent open-source projects such as microsoft/autogen, where he implemented teachable-agent features, memory persistence, and robust testing improvements. Based in Redmond, he brings rare institutional knowledge spanning speech, search, and modern agentic AI, plus a history of entrepreneurial and low-level systems work dating back to early game and DSP platforms. Colleagues rely on him to bridge ambitious research ideas into maintainable, scalable software.
10 years of coding experience
29 years of employment as a software developer
Bachelor’s Degree Electrical and Computer Engineering, Bachelor’s Degree Electrical and Computer Engineering at Brigham Young University
A programming framework for agentic AI 🤖 PyPi: autogen-agentchat Discord: https://aka.ms/autogen-discord Office Hour: https://aka.ms/autogen-officehour
Role in this project:
Full-stack Developer & ML Engineer
Contributions:172 reviews, 10 PRs, 245 pushes in 1 year 6 months
Contributions summary:Ricky made substantial contributions to the `microsoft/autogen` repository, focusing on the development and testing of the `TeachableAgent` functionality. Their work included implementing features related to agent teachability, response caching, memory persistence, and integrating the `TextAnalyzerAgent`. Furthermore, the user worked on adding unit tests and improving overall code quality, including refactoring, cleanup, and linter fixes. Additionally, their contributions involved modifications to the testing framework and installation instructions.
NAIL is an agent that plays text-based interactive fiction games.
Contributions:4 commits, 1 PR, 1 push in 3 years 3 months
agentgamesnailinteractive-fictiontext-based
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.