A few things I've built with AI.

A React Native prototype that analyzes one or more job-site photos and drafts hazards, mitigations, confidence ratings, and compliant conditions.

The model prepares a first pass. A worker reviews every finding, records corrections or dismissals, signs, and owns the final assessment.

The current evidence is the prototype's end-to-end review flow, which preserves the original model result alongside each human correction or dismissal.

Policy references are not yet grounded in an approved document store, and the model has not been evaluated against a reviewed set of job-site images.

A source-linked brief format and a draft workflow for reading several sources together.

The agent drafts. The reader decides what is credible and what deserves follow-up.

This build note is the public evidence for now. No runnable demo or evaluation is linked yet.

No evaluation set is published yet, so the project does not support reliability claims.

A structured brief schema and an early conversation flow for recording assumptions and open questions.

Missing evidence stays visible. The model does not declare the idea validated.

This build note is the public evidence for now. No runnable demo or user test is linked yet.

It still needs tests with people bringing their own incomplete ideas.

A comparison rubric organized around the job, cost, latency, and failure risk.

The recommendations show tradeoffs and uncertainty. They are not universal rankings.

This field note is the public artifact. No current benchmark results are linked yet.

Model behavior and pricing change often, so every recommendation needs a date and a repeatable test.