Stamped Energy: First Month of Enterprise Conversations

Vinayak Raizada and I pivoted to Stamped Energy in May 2026. We are not deployed in a plant yet. We are one month into conversations with plant managers, integrators, and people who have seen energy dashboards die in bookmark folders.

This post is what we are hearing before we claim any deployment wins.

How We Got Here

January 2026: Vinayak and I started Stamped (C2PA and image-authenticity, consumer ZK proofs, then B2B verification for logistics and insurance). Competitor mapping (Truepic and others). At ETHMumbai 2026 we won the Hashed Emergent VC track with the stamped-ethm ZK capture stack.

Mid-April 2026: we went to Bangalore for a YC event (April 15–20), explored the ecosystem, and talked to founders who had been building in India for years. That week changed how we thought about enterprise GTM.

By late spring, enterprise interviews kept surfacing a parallel pain: operational waste in manufacturing, terabytes of energy telemetry, and no decisions on top of it. In May 2026 we pivoted from Stamped to Stamped Energy. Same co-founder team. Different buyer and integration surface.

What Plant Managers Ask First

Not “show me your model.” Not “what is your Series A story.”

“Will this work with our historian without opening write access?” OPC-UA read-only is table stakes. If you need a firewall exception that takes six months, you are dead before you start.

“Can I trust the number on screen?” They have seen dashboards that disagree with the meter on the wall. Side-by-side historian cross-check is not a feature. It is admission requirements.

“Who maintains the tag mapping when equipment changes?” Integrators name tags like archaeology. Your product needs a UI plant staff will actually use, not a one-time professional services engagement.

“What happens when your model is wrong?” Maintenance calendars, dismiss-with-reason, and audit logs. False positives without context destroy trust faster than no product at all.

What We Are Building Toward

flowchart TB
    subgraph PlantOT["Plant OT Network"]
        PLC[PLCs / DCS]
        HIST[Plant Historian]
        OPC[OPC-UA Server]
    end

    subgraph DMZ["Site DMZ"]
        GW[Stamped Energy Gateway - Read Only]
        VAL[Validation Engine]
        QS[Quality Scorer]
    end

    subgraph Cloud["Stamped Energy Cloud"]
        TS[(Time Series Store)]
        ML[Anomaly Models]
        DASH[Energy Dashboard]
    end

    subgraph Users["Users"]
        PM[Plant Manager - War Room]
        ENG[Process Engineer - Drill Down]
    end

    PLC --> HIST --> OPC
    OPC -->|TLS, read-only| GW
    GW --> VAL --> QS
    QS --> TS
    TS --> ML --> DASH
    DASH --> PM
    DASH --> ENG

    style VAL fill:#1a3a5c,color:#fff
    style QS fill:#1a3a5c,color:#fff
    style PM fill:#2d5016,color:#fff

Architecture we are scoping with early conversations. Not shipped to production yet.

Month One Lessons

Data quality is the product before optimization is. Every serious conversation eventually lands on invalid readings, stale tags, and naming chaos. Intelligence on dirty data is a demo, not a deployment.

Credibility beats chart density. Plant managers want three numbers they believe, not twenty they ignore.

Sales cycles are measured in quarters, not sprints. We knew this from insurance discovery on Stamped. Energy confirms it with harder procurement and longer IT reviews.

A pivot is not a rebrand. Stamped was image authenticity. Stamped Energy is industrial energy intelligence. We kept the founder team and the discipline around trust; we did not pretend the SKUs are the same.

What Comes Next

More conversations. Honest scoping with plants willing to talk to an early team. No inflated deployment narrative until we have earned it.

If you are a plant manager who has buried an energy dashboard before, I want to hear what killed it. That feedback is worth more than any accelerator lecture.