Skip to content
View SethGemperline's full-sized avatar
  • Columbus, OH

Block or report SethGemperline

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.

Content in all repositories owned by your account will be closed.
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
SethGemperline/README.md

Seth Gemperline

Applied AI Systems Builder | Agentic AI | Human-AI Collaboration

I'm an independent AI systems builder focused on agent architectures, persistent memory, structured model workflows, and human-AI collaboration.

My work centers on a recurring question: How do we make AI systems capable of doing meaningful work over time without asking one giant model call to do everything?

Current Work

Project Mind Meld

A local-first human-AI collaboration environment for developing complex projects through natural conversation.

Mind Meld explores persistent project memory, bounded AI agents, explicit authority boundaries, context retrieval, structured model workflows, and the separation of AI judgment from deterministic system control.

Project Bellringer

An experimental game and AI research project exploring small-model autonomous agents, learned behavior, and natural-language tactical coaching.

The long-term research question is how much useful autonomous behavior can be achieved with very small models capable of running on constrained hardware.

Prior R&D

Project Crankshaft

A predecessor multi-agent research system exploring hierarchical decomposition, specialized agents, persistent retrieval, routing, synthesis, and incremental artifact production.

Lessons from Crankshaft directly informed the collaboration-first architecture of Project Mind Meld.

Artificial Judgment

Independent research into relational structure among embedded informational states, structural movement over time, and cross-embedding comparison.

This work resulted in four filed provisional patent applications.

Areas of Focus

  • Agentic AI systems
  • Human-AI collaboration
  • Persistent AI memory
  • LLM orchestration
  • Context retrieval and bounded context
  • AI evaluation and reliability
  • Tool-use and workflow design
  • Small-model experimentation
  • AI-assisted software development

Development Approach

My strongest work is in systems architecture, problem decomposition, workflow design, model behavior analysis, testing strategy, and defining the boundary between AI judgment and deterministic software authority.

I use AI coding agents extensively to implement and test the systems I design while continuing to develop my own programming proficiency.


Columbus, Ohio | Open to remote applied-AI opportunities

Pinned Loading

  1. project-mind-meld project-mind-meld Public

    Local-first human-AI collaboration system exploring durable project state, bounded AI reasoning, and deterministic authority.

  2. project-crankshaft project-crankshaft Public

    Experimental multi-agent AI research exploring hierarchical reasoning, persistent memory, retrieval, orchestration, and artifact production.

    Python

  3. artificial-judgment artificial-judgment Public

    Independent research into relational structure, structural change, and cross-embedding comparison in AI systems.