Software Designer with experience in UX, AI and Systems Level Architecture

Currently, my primary focus is in Next, EXLLaMa, and TRL. Recently, I've created Printer, a Next framework focused on being a productivity tool. However now, my work has shifted towards the creation of a RAG framework, featuring a training format inspired by the Alpaca Format but tailored for RAG optimization. This also includes a RAG testing suite for LLMs (better than MARCO for RAG). Coming soon, early 2024.

In the AI field, the concept of RAG (Retrieval-Augmented Generation) is a method for parsing, searching, and verifying data. However, a majority of the training and workflows within the LLM ecosystem is not optimized for it, this is obvious when dealing with larger context windows. I will be publishing a new framework in early 2024 that builds upon the Alpaca Format, offering enhanced capabilities to train both Base Models and Low Rank Adapters.

Right now, my focus is on developing turnkey LLM solutions, aimed at improving LLMs through advanced contextualizations and sophisticated native advertising. Asides from that, my most recent experience was working on a project called Open3, a qualified minting platform, where we collaborated on the creation of Liquid Death's NFT Collection Murder Head Death Club.


Automation Tooling for Next, Redux and Prisma

View Website

Hugging Face

TRL - Transformer Reinforcement Learning

TRL is a full stack library where we provide a set of tools to train transformer language models with Reinforcement Learning, from the Supervised Fine-tuning step (SFT), Reward Modeling step (RM) to the Proximal Policy Optimization (PPO) step. The library is integrated with 🤗 transformers.