Open-methods view · public synthetic data · open methods · no account needed
Sandbox · shadow-mode · not a supplier, aggregator or trading screen — the missing coordination layer.
INDICATIVE PROTOTYPE · figures pending NUMBERS v2.2 sign-off · internal — not for external release
Plane B · productDesign flex products
before you build them
Scoping question

Can a flex product be auditably fair to the consumer and viable for the supplier — before any customer is enrolled or the grid is touched?

Research inputs

Elexon-shaped half-hourly price weeks · Consumer-Duty winner/loser bands by segment · 5-gate stage-gate pack (concept → pilot).

Method

Backtest candidate tariffs on synthetic cohorts; deterministic core decides fairness, AI assists exploration only; shadow-mode throughout — the grid is never touched.

Flex Nexus · Lane B

Design and backtest flex products before you build them.

Prove to regulators they're auditably fair. Model how things could have been. Test the market rules before they're locked in — on real deterministic engines, empowered with AI at the fringe, on open data. indicative · synthetic cohort

Toy (free) → 2-week MVP (paid pilot) → Flex Nexus subscription → add-on modules.

Candidate · "Whole-home EV-TOU"synthetic
1 Concept
2 Back-test
3 Review
4 Shadow
5 Pilot
P&L · median
+ positive
downside · ES
banded
Auditably fair
Consumer-Duty card · winners/losers by segment
deterministic core decides fairness · AI assists exploration only
7 outputs

One backtest. Seven things you can take to a stage-gate.

P&L-under-uncertaintyconsumer-fairnessautomation-upliftreliabilitysettlement robustnessstage-gate packportfolio-fit
5 gates
concept → pilot stage-gate
deterministic core · shadow-mode
7 outputs
per backtest run
P&L · fairness · automation-uplift · reliability · settlement · stage-gate pack · portfolio-fit
2 weeks
paid-pilot MVP turnaround
indicative · subject to scoping