Michael Blacconiere
PHD Student at Indiana University Luddy School of Informatics, Computing, and Engineering
Atlanta, Georgia, United States
Summary
🤩
Rockstar🎓
Top SchoolMichael Blacconiere is a PhD student in Computer Engineering at Indiana University and a research engineer at Graf Research with 11 years of hands-on experience across embedded systems, HDL and physical circuit design, and full-stack software development. He bridges low-level hardware design and modern ML tooling, contributing to high-profile projects like huggingface/trl where he improved RL training integrations and performance techniques such as gradient checkpointing. Comfortable moving between research and product engineering, he has taught hardware description languages, led full-stack enterprise features at ServiceNow, and rapidly prototypes solutions for microelectronics challenges. Based in Atlanta, he pairs academic rigor with production-grade implementation skills and a knack for applying new concepts quickly to real-world hardware-software problems.
11 years of coding experience
1 year of employment as a software developer
Master of Science - MS, Computer Science, 3.6, Master of Science - MS, Computer Science, 3.6 at Indiana University Bloomington
Github Skills (10)
transformer-models10
huggingface-transformers10
pytorch10
nlp10
trainings10
python10
natural-language-processing10
modeling10
reinforcement-learning10
deepspeed8
Programming languages (9)
Github contributions (5)
huggingface/trl
Apr 2023 - Feb 2025
Train transformer language models with reinforcement learning.
Role in this project:
ML Engineer
ML EngineerContributions:19 reviews, 30 PRs, 86 comments in 1 year 10 months
Contributions summary:Michael primarily contributes to the training and integration of transformer language models with reinforcement learning within the trl library, as evident from the StackLLaMA and reward modeling integrations. Their work includes merging PEFT adapters, fixing RL training configurations, and improving reward modeling techniques, which are critical for the project's core functionality. Additionally, the user implemented gradient checkpointing and updated dependencies to incorporate the latest versions of the PEFT library.
language-modelreinforcement-learning
mnoukhov/dotfiles
Jan 2016 - May 2026
Contributions:32 pushes, 1 branch in 10 years 6 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.