by Antonio V. Franco

Measure what works.

Huyawo builds rigorous and reproducible foundations for evaluating efficient AI systems. Evidence comes before claims.

For projects and work: contact@antoniovfranco.com

Current focus

Evaluation before expansion.

Huyawo is concentrating its current work on the evaluation foundation required for later research programs. Public claims and releases follow verified evidence.

Active research and development

Huyawo Eval

A reproducible evaluation foundation for measured AI systems.

The work centers on explicit protocols, deterministic scoring, preserved raw results, provenance, remote validation, and disciplined release gates.

Research principles

Built for evidence.

01

Benchmarks before demonstrations

Research questions, baselines, metrics, costs, and failure criteria are defined before publicity.

02

Reproducibility before rankings

Configurations, revisions, raw results, and execution evidence matter more than isolated scores.

03

Limitations before release

Claims remain bounded by the evaluated conditions, with failures and unresolved risks preserved.

Partnerships and work

Start a serious conversation.

For research collaboration, engineering work, institutional partnerships, or open source contributions, contact Huyawo directly.