Building DPP Kit in Public: From Beta to Pilot to Production in 8 Weeks

· 6 min read
Building DPP Kit in Public: From Beta to Pilot to Production in 8 Weeks

Building DPP Kit in Public: From UNTP Compliance to Production in 4 Weeks

After a year managing UNTP pilot programs in the supply chain covering mines to OEMs, I kept hearing the same problem over and over:

"How to empower supply chain participants—from small suppliers to Fortune 500 manufacturers—to simply issue UNTP credentials?"

Meanwhile, the EU's ESPR deadline is approaching. Battery Passport regulations kick in 2027. Fortune 500 OEMs are mandating UNTP credentials from suppliers. The regulatory pressure is real, but the tools aren't keeping up.

So I'm building DPP Kit - production-ready UNTP credential infrastructure that organizations can deploy in days, not months.

And I'm doing it in public. Four weeks from beta launch to Vegas Tech Alley pitch. Here's the plan.


Week 1: Beta Launch (Feb 16) ✅

Shipping today: Core platform with Digital Product Passports (DPP) and Digital Facility Records (DFR).

What you can do:

  • Create your organization → get a did:web identifier automatically

    • Register as a Practitioner if you want to manage multiple organizations

  • Issue DPPs for your products

  • Issue DFRs for your manufacturing facilities

  • Link products to facilities (provenance chain)

  • Get QR codes for consumer verification

  • Public verification pages (scan and verify instantly)

Why start with just DPP and DFR? Because these two credential types cover 80% of EU ESPR compliance requirements. Get manufacturers issuing credentials now, add the rest later.

Beta is live: dppkit.io


Week 2: Automatic Conformance Validation (Feb 19-20)

What I'm building: Self-hosted UNTP Tier 2 validation pipeline.

Why it matters: Every credential you issue gets automatically validated against the official UNTP specification. No guessing whether you're compliant. You get a green checkmark ✅ or specific errors to fix.

How it works:

  • Pre-flight validation (blocks bad credentials before signing)

  • Post-issuance conformance check (proves it to auditors)

  • Public conformance badges on verification pages

Technical bit: Running our own validator instead of depending on the public test.uncefact.org endpoint. Better reliability, faster response, no external dependencies for production operations.

Status: Validation service already working. Integrating it into the full issuance pipeline.


Week 3: AI-Assisted Credential Creation (Feb 21-22) 🚀

The ambitious sprint: Two days to ship AI-powered credential creation.

What I'm building: Chat with Claude AI in plain English, get UNTP-compliant credentials.

Example:

You: "I need a DPP for battery-grade lithium carbonate, 99.5% purity, from our Nevada facility"

AI: "I'll create that. What's the production batch number?"

You: "LC-2026-001"

AI: [generates UNTP-compliant JSON]

System: [validates against Tier 2 specs]

AI: "Missing carbon footprint data - what's your extraction process emissions?"

You: "2.3 kg CO2e per kg lithium"

AI: [regenerates with sustainability claim]

System:Tier 2 Conformant

Why this is hard: The AI needs to understand UNTP schemas, generate valid JSON-LD, handle validation errors, and iteratively fix mistakes. All while feeling like a natural conversation.

Tech stack: Claude API + n8n workflows + UNTP validation loop. AI sees validator errors and corrects itself.

Is two days realistic? Probably not. But I'm going for it. Worst case it takes until Feb 25. Either way, it ships before month-end.


Week 4: Full UNTP Coverage (Feb 24-28)

What I'm adding:

  • Digital Traceability Events (DTE) - Supply chain events (manufacturing, shipment, transformation)

  • Digital Conformity Credentials (DCC) - Certifications and compliance attestations

  • Digital Identity Anchors (DIA) - Organizational identity verification

Why not launch with all five? Two reasons:

  1. DPP + DFR is enough for most pilots - Get users testing core features first

  2. DCC needs custom work - The UNTP spec for conformity credentials requires industry-specific vocabularies. DPP Kit DCC’s will require the collaboration of a UNTP practitioner and the standards/conformity organization that wishes to issue the DCC.

Reality check: DTE and DIA are straightforward (templates ready, just UI work). DCC will be released as a “pilot” version until the representing organization is satisfied with how the tool shows their conformity claims.


March 1-16: Polish and Vegas Prep

Two weeks of:

  • Bug fixes based on beta feedback

  • UX polish (AI chat needs to feel smooth)

  • Performance optimization

  • Vegas Tech Alley demo prep (March 17)

Goal: Walk into that pitch with 4 weeks of production usage, real user testimonials, and battle-tested AI features.


Why Build in Public?

Transparency: You'll see what works, what breaks, what takes longer than expected. This staged rollout (beta → production pilots → full SLA commitments) only works if we're learning from real usage at each phase.Feedback: Beta users shape the roadmap. If AI credential creation isn't useful, I'll pivot. If DIA is more critical than I thought, I'll prioritize it.

Accountability: I just committed to shipping AI in two days. Public pressure is motivating pressure.

Community: UNTP adoption needs an ecosystem. Building in public invites collaboration instead of competition.


Follow Along

I'll be posting updates as I build:

  • Progress updates - What shipped, what didn't, what I learned

  • Technical deep-dives - How the AI validation loop works, did:web architecture, conformance testing

  • Beta user stories - How organizations are using DPP Kit for real compliance work

Want to test the beta?Sign up at dppkit.io

Join the communityDPP Kit Zulip - Get help, share feedback, connect with other UNTP practitioners

Questions or feedback?Email me

Building something similar? → Let's talk. UNTP needs multiple implementations, not one monopoly.


What Could Go Wrong

Realistic risks I'm watching:

AI might suck: Two days to build AI-assisted creation could result in a clunky chatbot that doesn't understand UNTP well enough. Mitigation: Ship basic version, iterate based on feedback.

Beta users find critical bugs: Production systems always reveal edge cases. Mitigation: Cap beta at 30 organizations, fix issues fast.

DCC vocabulary problem: UNTP's conformity credential spec needs industry-specific extensions. Mitigation: Ship DPP/DFR/DTE first, build DCC template system properly in March.

I underestimated timelines: Two-day AI sprint might become five-day AI sprint. Mitigation: Vegas isn't until March 17. I have buffer.


Why Now

The window is closing:

EU ESPR implementation deadlines hit throughout 2026-2027. Organizations that wait for "ERP vendors to add UNTP support" will be 2+ years behind. Companies that act now get:

  • Early compliance advantage

  • Time to validate approach before deadlines

  • Competitive positioning with sustainability-focused buyers

The infrastructure gap:

The UN/CEFACT UNTP implementations register lists 17 software vendors. All 17 currently show status: "planned."

DPP Kit is bridging the gap between planning and production:

  • Today: Beta testing with early organizations

  • 4 weeks: Production-ready for pilot programs

  • 8 weeks: Full production tier with SLA commitments


Let's see what happens. 🚀

Building DPP Kit

Feb 16, 2026