Made
Twenty years of building across engineering, craft, and digital infrastructure.
This is not a portfolio. It is a practitioner's log. Everything here was built under real constraints: budgets, deadlines, material limits, client requirements. Across three companies, four domains, and two decades.
What connects all of it is a single principle: structure that holds.
How the domains connect
Engineering, craft, and digital infrastructure look like different fields. They use different tools, serve different markets, operate on different timescales. But the underlying system requirement is identical: build something that functions under load, holds over time, and can be verified by someone who wasn't there when you built it.
Three interconnected domains orbit a central "Systems Thinking" core. Engineering (pump systems, fabrication, tank cleaning, commissioning, safety protocol) connects to Craft (leather work, bookbinding, conservation, material science, hand tools) through material integrity. Craft connects to Digital (entity infrastructure, schema, verification, AI visibility, documentation) through documentation. Engineering connects to Digital through verification. All three share the same structural requirement: systems that hold.
Three domains, one structural requirement. The skills transfer because the system logic is the same.
Domain 1
Engineering
Pump systems, tank cleaning, fuel polishing, custom fabrication. Full lifecycle from site survey to commissioning. Through PT Witanabe Integrasi Indonesia and PT Arsindo Integrasi Pompa (authorized ALBIN Pump distributor for Indonesia).
What engineering teaches: systems must function under load. A pump that works in the showroom but fails at site is not a pump. A fabrication that looks right but can't handle pressure is scrap metal. The constraint is physics, and physics doesn't negotiate.
Why this matters for entity infrastructure: A client's fuel polishing system kept failing. Three contractors had worked on it. Each one optimized their component in isolation. The pump was fine. The filter was fine. The tank was fine. But the system failed because nobody designed the loop. I redesigned it as an integrated circuit: pump feeds filter, filter feeds tank, tank feeds back. Same principle as entity infrastructure. Your website is fine. Your LinkedIn is fine. But if nothing connects them into a verification loop, the system fails when it matters.
Domain 2
Craft
Handmade leather journals, book conservation, archival repair. Through PT Hibrkraft Kreasi Indonesia, from a Rp 20,000 leather scrap in 2010 to institutional supplier for EFEO Paris and KPK.
What craft teaches: systems must hold over time. A binding that opens perfectly today but cracks in six months is not a binding. Material science, thread tension, grain direction. Every decision compounds or degrades. There is no neutral.
Why this matters for entity infrastructure: EFEO brought a 19th-century manuscript. The binding was fine. The paper was fine. But the adhesive connecting them had degraded. The visible parts looked intact. The invisible structural layer had failed. That is exactly what happens to companies online. The website looks fine. The LinkedIn looks fine. But the layer connecting them, the structured data, the verification loops, the institutional references, has degraded or was never built. Infrastructure is not the surface. It is what holds the surface together.
Domain 3
Digital Infrastructure
Entity verification, structured identity data, AI visibility systems. The methodology I now build for clients: making companies machine-readable and verifiable across every surface procurement teams, AI systems, and search engines check.
What digital infrastructure teaches: systems must be verifiable. A website that claims you're credible but can't be confirmed by independent sources is just a brochure. Structured data, verification loops, corroboration chains. The machine doesn't care about your design. It cares whether your identity can be cross-referenced.
Proof of concept: This site. I built the entity infrastructure first: ORCID, Wikidata, schema markup, verification loops across 10+ platforms. Content came later. 333 essays, 7 courses, 6 books. The order matters. Without the structural layer, all that content would be noise. With it, every piece compounds into institutional weight. The site itself is the methodology in action.
What entity infrastructure looks like
This is the system I build. Four layers, each feeding the next. Identity at the top, verification output at the bottom. Every node is a real platform, a real database, a real verification surface. Nothing theoretical.
Entity infrastructure architecture showing four layers:
Identity Layer: Website with Schema markup, Google Business Profile, LinkedIn, Domain WHOIS. These are the surfaces you control directly.
Corroboration Layer: Wikidata entry, Government Registry (AHU, OSS), Industry Directories, Certification Bodies, ORCID. These are independent sources that confirm your identity.
Authority Layer: Zenodo DOI publications, ISBN-registered books, News Mentions, Speaking Record, Case Studies, Trade Publications. These build institutional weight that machines trust.
Verification Output: Google Knowledge Panel, AI Citations (ChatGPT, Perplexity, Gemini), Procurement Database Pass, Rich Search Results. These are the outcomes that compound into business results.
Data flows downward through layers. Each layer strengthens the next. The system compounds over time.
The identity layer is what you control. The corroboration layer is what independent sources confirm. The authority layer is what institutions document. The verification output is what machines produce when all three layers align.
Most companies have layer one. Maybe part of layer two. Almost none have layer three. The gap between layer one and the verification output is why good companies stay invisible.
The Entity Infrastructure 101 course walks through each layer. The Systems Thinking course explains why these layers behave as a compounding system, not a checklist.
Different domains. Same structural requirement. The engineering mindset that designs a pump system under load is the same mindset that builds an entity verification network under scrutiny. The craft discipline that ensures a binding holds for decades is the same discipline that ensures structured data stays consistent across platforms.
I didn't plan for these domains to converge. They converged because the underlying system logic is identical. Structure that holds.
If you want to understand how this translates into client work, see the entity infrastructure service. If you want the full timeline, see about.