A few things I’m building

Curiosity, put to work.

Products, open standards, and experiments. Each starts with a question about how people learn, build, or work with AI.

SayMakeLearning · Product & engineering

A place to become a builder.

AI-supported maker studio where young people build, test, explain, and improve projects.

Private preview

Web App

Visit saymake.app

Evidence & project status

Checked 14 October 2026

Problem: AI tools can do the thinking for kids.
What I built: A maker studio. Kids choose, build, test, and explain a project. The AI coaches. It does not build for them.
Status: No learner results published yet. Still unproven whether learners can show the skill without the AI.

Earlier capture of the SayMake maker studio

OMBSLearning · Open standards

What does it mean to build well?

Open Making and Building Standard: a draft, openly licensed standard describing the practices of making and building across K–12.

Experiment

Open Source

Read the draft standard

Evidence & project status

Checked 28 October 2026

The public repository and README describe draft v0.3.0, open for community review. Source access does not establish adoption, learning outcomes, or a validated assessment.

Open Making and Building StandardIdeas take shape.
Practice leaves evidence.Concept illustration · Draft standard

ReviewFlowAI tools · Human review

AI suggestions. Human decisions.

An experimental local review app: a model proposes labels, a person reviews and corrects them, and exports keep each decision, reason, and source.

Experiment

Open Source

View source

Evidence & project status

Checked 14 October 2026

Problem: Model proposals are suggestions, not ground truth.
What I built: A local review app. A model proposes labels. A person reviews and corrects them. Exports keep each decision, reason, and source.
Status: Public repo has setup and verification results; demo data is simulated. No real human benchmark yet.

ReviewFlowA person stays
in the loop.Concept diagram · Local experiment

SPEARAI security · Open source

Trust, checked at the boundary.

Secure Prompt Enforcement At Runtime: open-source LLM security middleware with a red-team evaluation harness.

Experiment

Open Source

View source

Evidence & project status

Checked 9 October 2026

Problem: AI app security is often asserted, not inspected.
What I built: Open-source LLM security middleware with a red-team evaluation harness.
Status: Public repo shows source, README, and evaluation method. Live models, new attacks, real integrations, output leakage, and latency remain unchecked.