Good Vibes & Thousands of Itterations

Vibe coding the Nexus suite.
Build. Measure. Learn.

I’ve been using AI-assisted development to build a connected suite of applications for creating, training, compressing, recovering, and running AI locally.

The work spans:

• Nexus Foundry — creating language models from random initialization.
• Nexus Compression & Recovery Studio — exploring smaller models and recovering capability after compression.
• Nexus Adapter Optimization Studio — developing specialized model adapters.
• Nexus Benchmark — measuring performance and identifying settings that can be applied to the actual applications.
• Nexus.cpp — developing my own inference engine, with a focus on Radeon hardware.

Alongside these, I’m working on Nexus Office, Imaging Studio, virtual machine tools, and the longer-term Nexus OS project.

My approach to vibe coding is hands-on: define the objective, work with AI to implement it, run the software, inspect the results, and bring the failures back into the next iteration.

A setting needs a working backend. A benchmark needs to represent the real workload. An exported model needs to load and produce useful results.

One recent milestone: Nexus Foundry reported approximately 24,893 training targets per second while training a 20-million-parameter model on a single AMD Radeon RX 7900 XT running Windows 11, using PyTorch.

That figure is a recent moving-average reading from a development run. Sustained performance and model quality remain part of the testing.

The process includes broken installers, stalled training runs, memory limits, and performance regressions. Each one gives me something concrete to investigate and improve.

That’s what I’m building toward with Nexus: a practical, connected environment for local AI development on hardware people already own.

Rich Dunbar | Nexus Emerging Technology

#VibeCoding #NexusSuite #AIAssistedDevelopment #LocalAI #PyTorch #AMD #Windows11 #BuildMeasureLearn