On AI Lounge since Oct 2026
Patrick Devaney
Member #699DeveloperL4Full Stack Developer, Independent
inference kernel optimizations, REAPs, fine tuning, ai agents, long horizon loops
How karma works. Everyone starts with 5. You get +1 each time another member upvotes one of your posts or comments (your own upvotes don't count), and lose it if they take the upvote back. Post what you're building, answer questions and help other members to earn more.
Achievements
Plugged In ×4Connected sources: GitHub, Hugging Face, LinkedIn, X, website.How: Connected 4 sources (GitHub, LinkedIn, X or a website) in Settings
Plugged In×4Featured
Shipper ×3Projects on your profile.How: Added 15 projects to their profile
Shipper×3
No posts yet
Patrick Devaney hasn't posted yet.
Pinned projects
mcp plugins to improve hermes agent for long-horizon engineering.
towards a faster, leaner web browser
Guide and resources on learning how to compile and use tensorrt-edge engines on the nvidia jetson agx thor
Model · 20,653 downloads · 19 likes
gguf
llama.cppreapmoeprunedendpoints_compatible
About
Patrick Devaney is an independent full stack developer in Miami who works on inference optimization, fine-tuning, and AI agents. He has worked in systems IT support at FIU since 2022. On GitHub he has published a guide to compiling TensorRT Edge LLM engines for Qwen models on the NVIDIA Jetson AGX Thor, and a demo of local agent workflows with Ollama and Swarms. Patrick is also building MCP plugins for long-horizon engineering with Hermes agents.
Categories
Stack
Experience
--
Systems IT Support · FIU
SD
PR Intern · Societa Dante Miami
Education
Florida International University
Continuing Education at the University of Miami
Uf College of Liberal Arts and Sciences
