Skip to content
View dunkeln's full-sized avatar
:octocat:
contemplating life
:octocat:
contemplating life

Block or report dunkeln

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 supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
dunkeln/README.md

dunkeln ๐ŸŒŒ

[หˆdสŠล‹kษ™ln] โ€” to grow dark (intransitive verb)

I build LLM tooling, agentic systems, and evaluation frameworks โ€” and explore deep learning research around emergent behavior and representation learning.


๐Ÿ”ญ Featured Work

Chaos Transformers โ€” Predicting Cellular Automata with Entropy-Based Patching
A two-stage transformer architecture modeling emergent stochastic behavior in 2D lattices using adaptive entropy-driven tokenization.
Read the blog โ†’


๐Ÿง  What I Do

  • Build reliable LLM systems (agents, RAG pipelines, evaluation tooling)
  • Design tokenization & patching schemes for structured data
  • Explore emergent phenomena in transformer architectures
  • Develop backend systems for ML-driven applications
  • Ship fast, minimalist full-stack products

๐Ÿ› ๏ธ Languages & Tools

Programming Languages

ML & Data

Backend / Infrastructure

Databases

Frontend


๐Ÿ“ฌ Connect


Pinned Loading

  1. transformer-stochastic-dynamics transformer-stochastic-dynamics Public

    Novel Autoregressive LM architecture predicting stochastic dynamics

    Python

  2. ragops ragops Public

    DX tool benching RAG Retrieval strategies

    Python

  3. llm-evals-lab llm-evals-lab Public

    Python

  4. trace_lm trace_lm Public

    Python