Skip to content
View avarga1's full-sized avatar

Block or report avarga1

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
avarga1/README.md

Austin Varga ⚡

Hi, I’m Austin —
I hate hard problems. I hate easy problems even more.


🚀 What I’m doing

  • Building Knowmi AI
  • Working on Rust-based LLM inference (vllm-hb)
  • Contributing to OSS (transformers, vllm, litellm, dotflow)
  • Building operation-reassurance (structural memory for LLMs)

🧠 Focus

  • Machine Learning Systems
  • Inference runtimes & schedulers
  • Agent / RAG architecture
  • Performance + infra

🔧 Tech

Python • Rust • CUDA • FastAPI • Postgres • SQLite • Vector DBs


📊 Systems footprint

Austin's GitHub stats Austin's most used languages

Austin's GitHub profile summary

Austin's GitHub activity graph


⚡ Current obsession

Designing systems that:

  • understand structure (not just tokens)
  • operate at runtime efficiently
  • make better decisions with context

📫 Contact

austinvarga1@protonmail.com

Pinned Loading

  1. vllm-hb vllm-hb Public

    vLLM-compatible inference runtime in pure Rust. Zero Python. Zero libtorch. CUDA via candle.

    Rust

  2. banana-standard banana-standard Public

    A peer-to-peer electronic fruit system. 1 BAN = 1 🍌. Forever. Nixon took us off gold. We're on bananas now.

    Dart

  3. dotflow dotflow Public

    Forked from dotflow-io/dotflow

    🎲 Dotflow turns an idea into flow! — Lightweight Python library for execution pipelines

    Python 1

  4. transformers transformers Public

    Forked from huggingface/transformers

    🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

    Python

  5. vllm vllm Public

    Forked from vllm-project/vllm

    A high-throughput and memory-efficient inference and serving engine for LLMs

    Python

  6. operation-reassurance operation-reassurance Public

    Repo health observatory. CST/AST-powered static analysis — test coverage, dead code, observability gaps, SOLID violations. No runtime required.

    Python 1 1