Curio Pattern Packet 0.1.0
The first packet passed an isolated two-agent reconstruction on a different learning topic. Public Gist publication remains separately approval-gated.
Dustin Persek · Solo builder
I turn practical needs and persistent questions into working systems, useful upstream changes, and evidence I can carry into the next thing.
· Active artifacts
A short ledger, updated when the underlying work changes.
The first packet passed an isolated two-agent reconstruction on a different learning topic. Public Gist publication remains separately approval-gated.
Reproduce first, make the smallest useful change, then test and review against the maintainers' rules.
The local operating stack behind parts of the work, shown here as supporting context rather than a prerequisite.
Primary wayfinding · Three territories
The same practice viewed from three directions. Each territory points to evidence, not a category shelf.
I make systems for real needs, use them, and keep the parts that survive contact with life.
I look for narrow public problems where careful reproduction, a maintainable patch, and good evidence can help.
First complete Pattern Packet
Curio turns a learning mission into bounded lessons, checks understanding, records evidence, and lets that evidence change the map.
The AI Systems topic is ongoing: seven lessons built so far, five quiz-passed, two ready but not yet reviewed, and one applied practice lab.
Field notes · Evidence before claims
Lessons, review receipts, and working artifacts make the practice inspectable.
A complete trace from mission and provisional map through a sourced lesson, visual poster, understanding check, evidence record, and the next map decision.
Inspect the evidence trailShipped proof · Selected work
These projects are supporting evidence of the practice, not a complete catalog.
Connection · About
I work best with builders who care about the evidence behind a claim, the boundary around a system, and the small implementation details that make it trustworthy.
I am especially interested in adaptive learning, agent-assisted development, maintainable open-source work, and tools that become more useful through lived practice.
More about my path