Selected work

AI systems delivered into commercially consequential energy workflows.

Blackdown works on the operational layer beneath the AI interface: pricing logic, optimisation, market integration, control boundaries and the evidence needed for people to approve the result.

Client identities and commercially sensitive methods are anonymised where required. Each case study states what was delivered, what was measured and where the architecture has been generalised.

Five process lines, each irregular until it reaches a single amber approval point, then straight
Architecture, build & deliveryA German power trading firm

Governed PPA, GoO and route-to-market workflows.

A governed agentic architecture for complex renewable-energy requests, with deterministic processing and explicit approval before binding actions.

Request triage reduced from approximately four hours to approximately twenty minutes in production.

A draft pricing line weaving around a straight reference line inside a narrow tolerance band, with the two out-of-band points marked in amber
AI consultingA European energy firm

Renewable origination pricing.

AI-assisted offer drafting with economics reconciled against desk reference pricing and exceptions surfaced for analyst review.

Against 24 historical deals, 78% of AI-generated proposals reconciled to analyst pricing within ±3%.

A price line spikes inside a highlighted window while a heavier state-of-charge line holds level through it; an amber segment beneath marks the reserved headroom explaining the hold
Reference architecture informed by delivered workMajor European utility

Governed BESS dispatch and revenue stacking.

Deterministic BESS optimisation, reserve-service integration and governed explanation around commercially consequential dispatch decisions.

Measured increase in BESS revenues. The precise uplift is commercially sensitive and is not published.

Bring us the workflow that still needs manual checking.