Published research

We publish what we learn.

Open-access papers by Josh Langsam, founder of Roanco Labs, on agentic AI, human–AI collaboration, and the memory architecture behind AI coding assistants. Every one carries a permanent DOI and is free to read. The same thinking goes into the systems we ship for clients.

Agentic AI

Breaking the Convergence Ceiling: Ten Observations from a Single Night

DOI 10.5281/zenodo.19547111 · open access Read the paper →

Human persistence

The Limit Rejection Principle: A Case Study of Human Persistence

DOI 10.5281/zenodo.19549560 · open access Read the paper →

Human–AI collaboration

Bandwidth Expanders: Human–AI Collaboration as Cognitive Channel Widening

DOI 10.5281/zenodo.19549863 · open access Read the paper →

Memory architecture

A Cognitive Memory Architecture for AI Coding Assistants

DOI 10.5281/zenodo.19549872 · open access Read the paper →

Research is not the product. The runtime it informs is. How we build and secure AI →

Book a call

90 minutes. One operator. A real roadmap on the table.

We’ll walk your operation, flag the workflows worth automating first, and outline what a Step 1 engagement would look like. No deck, no pressure, no commitment.

What you leave with
  • A hands-on walkthrough of your operation
  • The workflows worth automating first, ranked
  • What a Step 1 engagement would look like
Book a working session

Direct booking · replies within one business day · josh@roanco.com