← Back to portfolio

Lakehouse Marketplace · Interactive Prototype

Step 1 · Pick a dataset

You're a producer. Click any sample dataset below to load it. The platform auto-detects the schema and flags sensitive columns for the policy plane.

Step 2 · Define your SKU

Set the price, regions, clean-room requirement, and policy controls. The card on the right is a live preview of how your listing appears to consumers.

Listing
Pricing
Federation
Available regions consumers in any selected region can subscribe
Policy plane
Producer-approved queries only, optional differential-privacy outputs.
When off, queries from agents are rejected at the policy plane.
Sensitive columns (row/column-level controls)

Step 3 · Watch the marketplace work

Six simulated consumers will subscribe to your listing and run queries. The topology, revenue, and signed audit log update in real time. Policy violations (e.g. agents querying when agent access is off) are caught and logged.

Idle. Press play to start.

Live topology

human agent consortium · clean room policy violation

Live stats

Revenue
$0.00
Queries served
0
Active subs
0
Policy events
0

Revenue (live)

Signed audit trail · streams in real time

  • No events yet. Run the simulation to see the audit log fill in.

Step 4 · Recommendation, inputs, outputs

The platform's read on what just happened, the inputs that produced it, the processing variables, and standardized exports.

Recommendation

,

,

Key findings

    Top 3 next actions

      01Inputs · what you configured

      02Processing · what the platform did

      Revenue by consumer

      Query mix

      Region distribution

      03Outputs · the headline numbers

      Top consumers (by revenue)

      ConsumerTypeRegionQueriesViolationsRevenue

      04Standardized exports

      All artifacts download locally. Nothing leaves the page, this is a static prototype.